Apache Siege This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 633.35, N = 3 SE +/- 27.52, N = 3 SE +/- 8729.27, N = 12 SE +/- 310.67, N = 3 SE +/- 1450.97, N = 15 SE +/- 640.08, N = 3 SE +/- 590.10, N = 3 SE +/- 281.09, N = 3 36446 33372 122441 34460 53360 35979 35915 33989 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 35206.31 / Avg: 36446.01 / Max: 37291.17 Min: 33342.22 / Avg: 33371.94 / Max: 33426.93 Min: 33692.72 / Avg: 122440.73 / Max: 139586.83 Min: 33838.66 / Avg: 34459.9 / Max: 34780.19 Min: 43813.53 / Avg: 53360.21 / Max: 59438.89 Min: 34722.22 / Avg: 35978.82 / Max: 36818.85 Min: 34760.85 / Avg: 35915.13 / Max: 36705.33 Min: 33426.93 / Avg: 33989.01 / Max: 34279.45 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 58.16, N = 3 SE +/- 534.46, N = 7 SE +/- 368.77, N = 14 SE +/- 404.41, N = 9 SE +/- 399.98, N = 15 SE +/- 70.88, N = 3 SE +/- 466.62, N = 8 SE +/- 7609.36, N = 12 33412 43132 40546 36344 39870 33428 39959 122101 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 33327.78 / Avg: 33411.5 / Max: 33523.3 Min: 40056.08 / Avg: 43132.06 / Max: 44533.51 Min: 36403.35 / Avg: 40545.96 / Max: 42149.63 Min: 33422.46 / Avg: 36343.73 / Max: 37601.05 Min: 35784.58 / Avg: 39870.46 / Max: 41657.99 Min: 33338.89 / Avg: 33428.35 / Max: 33568.31 Min: 37037.04 / Avg: 39959.19 / Max: 40975.21 Min: 39062.5 / Avg: 122100.72 / Max: 133155.78 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 42.99 42.97 42.93 42.94 25.71 25.73 25.71 25.72 1. (CC) gcc options: -lm -lpthread -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 Min: 42.98 / Avg: 42.99 / Max: 43 Min: 42.96 / Avg: 42.97 / Max: 42.99 Min: 42.92 / Avg: 42.93 / Max: 42.95 Min: 42.91 / Avg: 42.94 / Max: 42.97 Min: 25.7 / Avg: 25.71 / Max: 25.73 Min: 25.72 / Avg: 25.73 / Max: 25.74 Min: 25.7 / Avg: 25.71 / Max: 25.73 Min: 25.71 / Avg: 25.72 / Max: 25.74 1. (CC) gcc options: -lm -lpthread -O3 -march=native -pipe
Timed ImageMagick Compilation This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 15.38 15.58 10.55 10.55 17.03 17.22 13.03 13.13 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 15.31 / Avg: 15.38 / Max: 15.5 Min: 15.56 / Avg: 15.58 / Max: 15.62 Min: 10.48 / Avg: 10.55 / Max: 10.68 Min: 10.38 / Avg: 10.55 / Max: 10.76 Min: 16.82 / Avg: 17.03 / Max: 17.16 Min: 17.02 / Avg: 17.22 / Max: 17.36 Min: 13.01 / Avg: 13.03 / Max: 13.06 Min: 13.03 / Avg: 13.13 / Max: 13.26
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 4.91, N = 3 SE +/- 2.70, N = 3 SE +/- 3.27, N = 3 SE +/- 1.23, N = 3 SE +/- 1.24, N = 3 SE +/- 1.89, N = 3 SE +/- 1.04, N = 3 SE +/- 1.19, N = 3 2387 2386 2354 1963 2515 2612 1703 2308 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 Min: 2377.12 / Avg: 2386.62 / Max: 2393.54 Min: 2381.14 / Avg: 2385.82 / Max: 2390.51 Min: 2347.39 / Avg: 2353.94 / Max: 2357.33 Min: 1960.16 / Avg: 1962.51 / Max: 1964.31 Min: 2512.96 / Avg: 2515.27 / Max: 2517.22 Min: 2608.58 / Avg: 2611.66 / Max: 2615.1 Min: 1701.29 / Avg: 1702.72 / Max: 1704.73 Min: 2305.68 / Avg: 2308.06 / Max: 2309.42 1. (CC) gcc options: -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 SE +/- 1.30, N = 3 SE +/- 0.35, N = 3 SE +/- 3.43, N = 3 SE +/- 0.10, N = 3 SE +/- 4.56, N = 3 SE +/- 10.69, N = 3 SE +/- 5.38, N = 3 SE +/- 0.37, N = 3 5639 5643 5682 4051 5603 5913 5174 3982 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 1000 2000 3000 4000 5000 Min: 5636.23 / Avg: 5638.62 / Max: 5640.72 Min: 5642.25 / Avg: 5642.93 / Max: 5643.42 Min: 5674.91 / Avg: 5681.6 / Max: 5686.3 Min: 4050.85 / Avg: 4051.01 / Max: 4051.2 Min: 5596.56 / Avg: 5602.82 / Max: 5611.7 Min: 5901.21 / Avg: 5912.88 / Max: 5934.22 Min: 5166.31 / Avg: 5173.79 / Max: 5184.22 Min: 3980.91 / Avg: 3981.57 / Max: 3982.18 1. (CC) gcc options: -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 1567 1568 1565 1564 2223 2224 2223 2222 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 1566.91 / Avg: 1566.95 / Max: 1566.98 Min: 1567.95 / Avg: 1567.97 / Max: 1568 Min: 1564.3 / Avg: 1564.56 / Max: 1564.69 Min: 1563.96 / Avg: 1564.02 / Max: 1564.07 Min: 2223.05 / Avg: 2223.29 / Max: 2223.45 Min: 2223.95 / Avg: 2224.01 / Max: 2224.05 Min: 2222.76 / Avg: 2222.8 / Max: 2222.87 Min: 2221.67 / Avg: 2221.96 / Max: 2222.28 1. (CC) gcc options: -march=native -pipe -O3 -lm
BYTE Unix Benchmark This is a test of BYTE. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 GRAPHITE LTO LTO GRAPHITE 13M 26M 39M 52M 65M SE +/- 245871.68, N = 3 SE +/- 197747.51, N = 3 SE +/- 177886.75, N = 3 SE +/- 535439.20, N = 3 SE +/- 77816.46, N = 3 SE +/- 128858.07, N = 3 SE +/- 266227.30, N = 3 SE +/- 114177.08, N = 3 45080736 43930746 59872198 60717315 45017240 44106377 59847981 59940786 1. (CC) gcc options: -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 GRAPHITE LTO LTO GRAPHITE 11M 22M 33M 44M 55M Min: 44634874.7 / Avg: 45080736.2 / Max: 45483289.6 Min: 43539494.5 / Avg: 43930745.63 / Max: 44176412.6 Min: 59595555.3 / Avg: 59872198.13 / Max: 60204253 Min: 59735463.3 / Avg: 60717314.63 / Max: 61578458.3 Min: 44865295.5 / Avg: 45017240.33 / Max: 45122380.9 Min: 43858924.8 / Avg: 44106377.03 / Max: 44292463 Min: 59539894.4 / Avg: 59847981.03 / Max: 60378113.3 Min: 59783361.3 / Avg: 59940786.07 / Max: 60162754.5 1. (CC) gcc options: -march=native -pipe -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 13.50, N = 3 SE +/- 54.34, N = 3 SE +/- 167.49, N = 3 SE +/- 82.50, N = 3 SE +/- 105.73, N = 3 SE +/- 90.50, N = 3 SE +/- 410.00, N = 12 SE +/- 101.00, N = 3 15078 14971 12340 12624 15055 15002 11350 11907 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 15051 / Avg: 15078 / Max: 15092 Min: 14870 / Avg: 14971.33 / Max: 15056 Min: 12010 / Avg: 12339.67 / Max: 12556 Min: 12459 / Avg: 12624 / Max: 12708 Min: 14857 / Avg: 15055.33 / Max: 15218 Min: 14821 / Avg: 15002 / Max: 15093 Min: 7144.8 / Avg: 11350.07 / Max: 12187 Min: 11705 / Avg: 11907 / Max: 12008 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 SE +/- 6.31, N = 3 SE +/- 5.13, N = 3 SE +/- 3.43, N = 3 SE +/- 3.65, N = 3 SE +/- 4.42, N = 3 SE +/- 5.30, N = 3 SE +/- 1.77, N = 3 SE +/- 2.27, N = 3 3198 3183 2915 2638 3198 3355 2692 2641 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3187.2 / Avg: 3198.24 / Max: 3209.07 Min: 3174.06 / Avg: 3183.27 / Max: 3191.78 Min: 2911.47 / Avg: 2915.41 / Max: 2922.24 Min: 2633.76 / Avg: 2637.82 / Max: 2645.11 Min: 3189.35 / Avg: 3198.16 / Max: 3203.18 Min: 3349.41 / Avg: 3355.12 / Max: 3365.7 Min: 2688.63 / Avg: 2692.08 / Max: 2694.48 Min: 2637.23 / Avg: 2640.63 / Max: 2644.93 1. (CC) gcc options: -march=native -pipe -O3 -lm
LLVM Test Suite This test times how long it takes to run the LLVM Test Suite. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better LLVM Test Suite 6.0.0 Time To Run GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 SE +/- 0.27, N = 3 SE +/- 0.10, N = 3 SE +/- 0.18, N = 3 SE +/- 0.29, N = 3 SE +/- 0.19, N = 3 SE +/- 0.32, N = 3 SE +/- 0.15, N = 3 SE +/- 0.28, N = 3 177 189 185 216 189 188 183 220 1. (CXX) g++ options: -march=native -pipe -O3 -lstdc++ Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better LLVM Test Suite 6.0.0 Time To Run GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 Min: 176.33 / Avg: 176.7 / Max: 177.23 Min: 189.36 / Avg: 189.46 / Max: 189.66 Min: 184.25 / Avg: 184.52 / Max: 184.85 Min: 215.21 / Avg: 215.74 / Max: 216.21 Min: 188.45 / Avg: 188.68 / Max: 189.05 Min: 187.05 / Avg: 187.55 / Max: 188.15 Min: 182.54 / Avg: 182.84 / Max: 183.04 Min: 219.31 / Avg: 219.64 / Max: 220.21 1. (CXX) g++ options: -march=native -pipe -O3 -lstdc++
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: List Suites GRAPHITE LTO LTO GRAPHITE 0.1913 0.3826 0.5739 0.7652 0.9565 SE +/- 0.01, N = 2 SE +/- 0.15, N = 2 SE +/- 0.02, N = 2 SE +/- 0.00, N = 2 SE +/- 0.02, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.02, N = 2 0.72 0.85 0.72 0.73 0.77 0.69 0.70 0.71 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: List Suites GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 0.71 / Avg: 0.72 / Max: 0.73 Min: 0.7 / Avg: 0.85 / Max: 1 Min: 0.7 / Avg: 0.72 / Max: 0.74 Min: 0.72 / Avg: 0.73 / Max: 0.73 Min: 0.74 / Avg: 0.77 / Max: 0.79 Min: 0.68 / Avg: 0.69 / Max: 0.69 Min: 0.7 / Avg: 0.7 / Max: 0.7 Min: 0.69 / Avg: 0.71 / Max: 0.72
ebizzy This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 GRAPHITE LTO LTO GRAPHITE 130K 260K 390K 520K 650K SE +/- 8631.83, N = 4 SE +/- 7283.08, N = 3 SE +/- 4747.42, N = 3 SE +/- 6383.63, N = 10 SE +/- 9659.20, N = 3 SE +/- 6097.02, N = 15 SE +/- 5616.97, N = 3 SE +/- 2130.51, N = 3 554576 530223 586533 592272 484539 526045 587332 583984 1. (CC) gcc options: -pthread -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 GRAPHITE LTO LTO GRAPHITE 100K 200K 300K 400K 500K Min: 532511 / Avg: 554575.75 / Max: 574618 Min: 521096 / Avg: 530223.33 / Max: 544618 Min: 580182 / Avg: 586533 / Max: 595821 Min: 548465 / Avg: 592271.9 / Max: 619100 Min: 465286 / Avg: 484539 / Max: 495541 Min: 489152 / Avg: 526045 / Max: 574404 Min: 577414 / Avg: 587332.33 / Max: 596860 Min: 579723 / Avg: 583984 / Max: 586123 1. (CC) gcc options: -pthread -lpthread -march=native -pipe -O3
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.B GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 11.86, N = 3 SE +/- 4.94, N = 3 SE +/- 28.38, N = 3 SE +/- 16.91, N = 3 SE +/- 18.65, N = 3 SE +/- 7.00, N = 3 SE +/- 23.06, N = 3 SE +/- 22.95, N = 3 15275 15232 12658 12623 15292 15278 12652 12588 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.B GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 15259 / Avg: 15275.01 / Max: 15298.18 Min: 15222.6 / Avg: 15232.47 / Max: 15237.93 Min: 12619.74 / Avg: 12658.03 / Max: 12713.47 Min: 12591.13 / Avg: 12622.78 / Max: 12648.92 Min: 15267.99 / Avg: 15292.4 / Max: 15329.04 Min: 15264.12 / Avg: 15277.95 / Max: 15286.75 Min: 12616.76 / Avg: 12651.83 / Max: 12695.3 Min: 12564.89 / Avg: 12587.85 / Max: 12633.75 1. 3.3
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.A GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 3.22, N = 3 SE +/- 75.72, N = 3 SE +/- 86.79, N = 3 SE +/- 83.26, N = 3 SE +/- 106.53, N = 3 SE +/- 116.83, N = 3 SE +/- 195.31, N = 3 SE +/- 16.67, N = 3 14441 14352 12075 12090 14275 14301 11916 12175 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.A GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 14436.84 / Avg: 14441.25 / Max: 14447.51 Min: 14200.61 / Avg: 14351.88 / Max: 14433.8 Min: 11907.6 / Avg: 12074.66 / Max: 12198.99 Min: 11930.65 / Avg: 12089.51 / Max: 12212.17 Min: 14101.6 / Avg: 14274.77 / Max: 14468.84 Min: 14067.58 / Avg: 14301.1 / Max: 14425.08 Min: 11554.22 / Avg: 11915.94 / Max: 12224.49 Min: 12151.38 / Avg: 12175.17 / Max: 12207.28 1. 3.3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 90K 180K 270K 360K 450K SE +/- 7010.78, N = 3 SE +/- 5194.21, N = 15 SE +/- 6644.14, N = 15 SE +/- 4273.23, N = 3 SE +/- 4305.07, N = 3 SE +/- 5248.76, N = 3 SE +/- 7556.40, N = 3 SE +/- 5594.88, N = 12 347632 403580 397741 408492 398950 403712 398632 401979 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 70K 140K 210K 280K 350K Min: 338177.02 / Avg: 347632.49 / Max: 361326.75 Min: 348062.54 / Avg: 403580.32 / Max: 440619.62 Min: 337953.8 / Avg: 397741.18 / Max: 439484.98 Min: 400000 / Avg: 408492.32 / Max: 413570.28 Min: 391437.31 / Avg: 398950.4 / Max: 406349.21 Min: 393241.17 / Avg: 403711.54 / Max: 409600 Min: 383520.6 / Avg: 398632.26 / Max: 406349.21 Min: 345712.36 / Avg: 401979.3 / Max: 413904.61 1. (CC) gcc options: -O2
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 48 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 42.71, N = 3 SE +/- 78.56, N = 3 SE +/- 6.49, N = 3 SE +/- 8.21, N = 3 SE +/- 9.69, N = 3 SE +/- 11.02, N = 3 SE +/- 165.12, N = 9 SE +/- 7.48, N = 3 11282 9808 11335 11366 11312 9748 11143 11437 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 48 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 11203.2 / Avg: 11282.43 / Max: 11349.7 Min: 9713.88 / Avg: 9808.15 / Max: 9964.15 Min: 11325.6 / Avg: 11334.77 / Max: 11347.3 Min: 11352.6 / Avg: 11365.93 / Max: 11380.9 Min: 11300.8 / Avg: 11311.9 / Max: 11331.2 Min: 9736.73 / Avg: 9747.99 / Max: 9770.02 Min: 9875.09 / Avg: 11142.72 / Max: 11365.7 Min: 11424.4 / Avg: 11437.33 / Max: 11450.3 1. (CC) gcc options: -lpopt -march=native -pipe -O3
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 12 GRAPHITE LTO LTO GRAPHITE 1500 3000 4500 6000 7500 SE +/- 101.37, N = 9 SE +/- 6.66, N = 3 SE +/- 6.42, N = 3 SE +/- 2.66, N = 3 SE +/- 1.30, N = 3 SE +/- 2.96, N = 3 SE +/- 5.15, N = 3 SE +/- 4.07, N = 3 6614 6071 7042 6854 6803 6059 6174 6962 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 12 GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 Min: 6081.7 / Avg: 6614.34 / Max: 6797.43 Min: 6063.61 / Avg: 6070.59 / Max: 6083.9 Min: 7028.81 / Avg: 7041.65 / Max: 7048.17 Min: 6849.32 / Avg: 6854.38 / Max: 6858.33 Min: 6800.45 / Avg: 6802.71 / Max: 6804.96 Min: 6054.7 / Avg: 6059.25 / Max: 6064.8 Min: 6164.39 / Avg: 6174.46 / Max: 6181.36 Min: 6953.98 / Avg: 6962.09 / Max: 6966.83 1. (CC) gcc options: -lpopt -march=native -pipe -O3
Apache Siege This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 10 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 814.60, N = 4 SE +/- 846.31, N = 3 SE +/- 750.12, N = 5 SE +/- 682.09, N = 6 SE +/- 602.01, N = 8 SE +/- 611.16, N = 8 SE +/- 573.84, N = 15 SE +/- 1018.25, N = 3 51320 47349 49455 49842 49284 49407 49669 54782 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 10 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 50505.05 / Avg: 51319.65 / Max: 53763.44 Min: 45662.1 / Avg: 47349.4 / Max: 48309.18 Min: 46511.62 / Avg: 49454.59 / Max: 50505.05 Min: 46511.62 / Avg: 49841.63 / Max: 51020.41 Min: 46511.62 / Avg: 49283.86 / Max: 50761.42 Min: 46082.95 / Avg: 49407.23 / Max: 50761.42 Min: 44444.45 / Avg: 49669.32 / Max: 52910.05 Min: 53763.44 / Avg: 54781.69 / Max: 56818.18 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 0.33, N = 3 SE +/- 399.05, N = 15 SE +/- 177.14, N = 13 SE +/- 258.83, N = 3 SE +/- 267.46, N = 5 SE +/- 253.18, N = 15 SE +/- 228.25, N = 8 SE +/- 161.91, N = 3 18018 16166 18536 17909 18269 17876 18653 17897 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 18017 / Avg: 18017.67 / Max: 18018 Min: 13556 / Avg: 16165.73 / Max: 17909 Min: 17756 / Avg: 18535.85 / Max: 19362 Min: 17391 / Avg: 17908.67 / Max: 18168 Min: 17994 / Avg: 18269.2 / Max: 19339 Min: 15690 / Avg: 17875.8 / Max: 19361 Min: 17889 / Avg: 18653.25 / Max: 19305 Min: 17626 / Avg: 17897 / Max: 18186 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.18 Time To Compile GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 SE +/- 0.50, N = 4 SE +/- 0.59, N = 3 SE +/- 0.60, N = 4 SE +/- 0.63, N = 4 SE +/- 0.48, N = 5 SE +/- 0.67, N = 3 SE +/- 0.50, N = 5 SE +/- 0.55, N = 4 33.67 33.91 38.76 38.64 33.65 33.97 34.04 34.26 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.18 Time To Compile GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 Min: 33.06 / Avg: 33.67 / Max: 35.16 Min: 33.3 / Avg: 33.91 / Max: 35.09 Min: 37.68 / Avg: 38.76 / Max: 40.47 Min: 37.47 / Avg: 38.64 / Max: 40.41 Min: 32.78 / Avg: 33.65 / Max: 35.5 Min: 33.26 / Avg: 33.97 / Max: 35.31 Min: 33.2 / Avg: 34.04 / Max: 35.99 Min: 33.59 / Avg: 34.26 / Max: 35.92
Timed MrBayes Analysis This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.1.2 Primate Phylogeny Analysis GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.17, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 50.65 58.18 50.53 58.18 50.73 58.14 50.54 57.96 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.1.2 Primate Phylogeny Analysis GRAPHITE LTO LTO GRAPHITE 12 24 36 48 60 Min: 50.5 / Avg: 50.65 / Max: 50.73 Min: 58.05 / Avg: 58.18 / Max: 58.39 Min: 50.43 / Avg: 50.53 / Max: 50.66 Min: 57.98 / Avg: 58.18 / Max: 58.51 Min: 50.65 / Avg: 50.73 / Max: 50.84 Min: 58.11 / Avg: 58.14 / Max: 58.2 Min: 50.37 / Avg: 50.54 / Max: 50.64 Min: 57.92 / Avg: 57.96 / Max: 57.98
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack GRAPHITE LTO LTO GRAPHITE 1.0305 2.061 3.0915 4.122 5.1525 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 4.14 4.05 4.52 4.48 4.05 4.14 4.58 4.50 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 4.14 / Avg: 4.14 / Max: 4.14 Min: 4.04 / Avg: 4.05 / Max: 4.06 Min: 4.51 / Avg: 4.52 / Max: 4.54 Min: 4.48 / Avg: 4.48 / Max: 4.48 Min: 4.04 / Avg: 4.05 / Max: 4.06 Min: 4.13 / Avg: 4.14 / Max: 4.14 Min: 4.54 / Avg: 4.58 / Max: 4.6 Min: 4.49 / Avg: 4.5 / Max: 4.5 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 200K 400K 600K 800K 1000K SE +/- 7777.06, N = 12 SE +/- 14259.77, N = 3 SE +/- 15504.31, N = 3 SE +/- 8295.96, N = 11 SE +/- 15609.75, N = 15 SE +/- 7388.33, N = 3 SE +/- 4500.72, N = 3 SE +/- 7801.18, N = 3 777719 740082 820873 793503 771040 836048 802768 803290 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 140K 280K 420K 560K 700K Min: 717086.84 / Avg: 777719.21 / Max: 808208.37 Min: 713588.85 / Avg: 740082.32 / Max: 762472.08 Min: 798129.38 / Avg: 820872.73 / Max: 850498.34 Min: 716585.02 / Avg: 793503.46 / Max: 824476.65 Min: 654731.46 / Avg: 771039.56 / Max: 854045.04 Min: 824476.65 / Avg: 836048.47 / Max: 849792.53 Min: 796267.5 / Avg: 802767.82 / Max: 811410.46 Min: 787692.31 / Avg: 803290.26 / Max: 811410.46 1. (CC) gcc options: -O2
dcraw This test times how long it takes to convert several high-resolution RAW NEF image files to PPM image format using dcraw. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better dcraw RAW To PPM Image Conversion GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 11.48 11.70 11.46 11.11 10.36 10.42 10.55 10.47 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better dcraw RAW To PPM Image Conversion GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 11.45 / Avg: 11.48 / Max: 11.51 Min: 11.64 / Avg: 11.7 / Max: 11.76 Min: 11.42 / Avg: 11.46 / Max: 11.5 Min: 11.08 / Avg: 11.11 / Max: 11.13 Min: 10.33 / Avg: 10.36 / Max: 10.39 Min: 10.42 / Avg: 10.42 / Max: 10.44 Min: 10.53 / Avg: 10.55 / Max: 10.58 Min: 10.45 / Avg: 10.47 / Max: 10.49 1. (CC) gcc options: -march=native -pipe -O3 -lm
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 11.55, N = 3 SE +/- 54.60, N = 3 SE +/- 303.17, N = 15 SE +/- 242.90, N = 15 SE +/- 53.80, N = 3 SE +/- 62.81, N = 3 SE +/- 8.39, N = 3 SE +/- 51.35, N = 3 12107 11795 11140 10796 11919 11871 11341 11291 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 12087 / Avg: 12107.33 / Max: 12127 Min: 11710 / Avg: 11795.33 / Max: 11897 Min: 8212.2 / Avg: 11140.17 / Max: 12231 Min: 9025 / Avg: 10796.08 / Max: 12060 Min: 11816 / Avg: 11919.33 / Max: 11997 Min: 11747 / Avg: 11871.33 / Max: 11949 Min: 11326 / Avg: 11341 / Max: 11355 Min: 11188 / Avg: 11290.67 / Max: 11344 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Info Command GRAPHITE LTO LTO GRAPHITE 0.3555 0.711 1.0665 1.422 1.7775 SE +/- 0.00, N = 2 SE +/- 0.09, N = 2 SE +/- 0.00, N = 2 SE +/- 0.10, N = 2 SE +/- 0.00, N = 2 SE +/- 0.07, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 1.41 1.56 1.47 1.58 1.47 1.53 1.46 1.47 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Info Command GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.41 / Avg: 1.41 / Max: 1.41 Min: 1.47 / Avg: 1.56 / Max: 1.65 Min: 1.47 / Avg: 1.47 / Max: 1.47 Min: 1.48 / Avg: 1.58 / Max: 1.68 Min: 1.47 / Avg: 1.47 / Max: 1.47 Min: 1.47 / Avg: 1.53 / Max: 1.6 Min: 1.46 / Avg: 1.46 / Max: 1.46 Min: 1.47 / Avg: 1.47 / Max: 1.47
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 130 260 390 520 650 SE +/- 6.48, N = 3 SE +/- 7.68, N = 15 SE +/- 7.82, N = 6 SE +/- 7.32, N = 15 SE +/- 1.74, N = 3 SE +/- 10.17, N = 3 SE +/- 8.89, N = 3 SE +/- 7.56, N = 15 579 580 570 579 582 592 530 573 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 100 200 300 400 500 Min: 567.01 / Avg: 579.2 / Max: 589.12 Min: 508.46 / Avg: 580.35 / Max: 615.7 Min: 535.21 / Avg: 569.5 / Max: 590.03 Min: 524.47 / Avg: 578.53 / Max: 625.79 Min: 578.56 / Avg: 581.8 / Max: 584.5 Min: 574.24 / Avg: 592.23 / Max: 609.46 Min: 516.95 / Avg: 529.5 / Max: 546.67 Min: 527.91 / Avg: 573.35 / Max: 617.88 1. (CC) gcc options: -O2
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 158.29, N = 3 SE +/- 108.31, N = 3 SE +/- 55.60, N = 3 SE +/- 88.42, N = 3 SE +/- 56.16, N = 3 SE +/- 21.99, N = 3 SE +/- 58.35, N = 3 SE +/- 188.43, N = 3 10674 10759 11611 11500 10893 10960 11845 11169 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 10361 / Avg: 10673.67 / Max: 10873 Min: 10568 / Avg: 10759 / Max: 10943 Min: 11508 / Avg: 11610.67 / Max: 11699 Min: 11400 / Avg: 11499.67 / Max: 11676 Min: 10784 / Avg: 10893 / Max: 10971 Min: 10916 / Avg: 10959.67 / Max: 10986 Min: 11746 / Avg: 11845 / Max: 11948 Min: 10922 / Avg: 11169 / Max: 11539 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 6 GRAPHITE LTO LTO GRAPHITE 1000 2000 3000 4000 5000 SE +/- 7.87, N = 3 SE +/- 5.44, N = 3 SE +/- 5.52, N = 3 SE +/- 75.93, N = 3 SE +/- 14.83, N = 3 SE +/- 2.06, N = 3 SE +/- 2.26, N = 3 SE +/- 7.00, N = 3 4052 4074 4484 4163 4062 4056 4117 4423 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 6 GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 Min: 4039.97 / Avg: 4051.76 / Max: 4066.69 Min: 4067.08 / Avg: 4073.95 / Max: 4084.69 Min: 4475.03 / Avg: 4483.85 / Max: 4494.01 Min: 4085.2 / Avg: 4162.83 / Max: 4314.67 Min: 4034.66 / Avg: 4062.08 / Max: 4085.58 Min: 4053.26 / Avg: 4055.95 / Max: 4059.99 Min: 4112.24 / Avg: 4116.6 / Max: 4119.78 Min: 4415.98 / Avg: 4423.27 / Max: 4437.27 1. (CC) gcc options: -lpopt -march=native -pipe -O3
Tinymembench This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 154.87, N = 3 SE +/- 556.96, N = 3 SE +/- 1871.91, N = 3 SE +/- 318.81, N = 3 SE +/- 222.91, N = 3 SE +/- 342.53, N = 3 SE +/- 1504.62, N = 3 SE +/- 109.36, N = 3 22147 22098 23526 21323 22609 21522 23449 22120 1. (CC) gcc options: -O2 -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 21897 / Avg: 22146.83 / Max: 22430.3 Min: 21000.2 / Avg: 22098.13 / Max: 22810 Min: 21309.5 / Avg: 23525.9 / Max: 27247.1 Min: 20688.1 / Avg: 21323.33 / Max: 21688.7 Min: 22344.3 / Avg: 22609.2 / Max: 23052.2 Min: 20839.9 / Avg: 21522.17 / Max: 21916.9 Min: 21530.3 / Avg: 23449.4 / Max: 26416.3 Min: 21912.1 / Avg: 22119.8 / Max: 22283 1. (CC) gcc options: -O2 -march=native -pipe -O3 -lm
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Ctype GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.10, N = 3 SE +/- 0.01, N = 3 SE +/- 0.48, N = 15 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.52, N = 14 SE +/- 0.01, N = 3 31.74 31.94 34.27 34.96 31.75 31.89 34.90 34.28 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Ctype GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 Min: 31.73 / Avg: 31.74 / Max: 31.75 Min: 31.74 / Avg: 31.94 / Max: 32.04 Min: 34.26 / Avg: 34.27 / Max: 34.29 Min: 34.26 / Avg: 34.96 / Max: 41.41 Min: 31.73 / Avg: 31.75 / Max: 31.77 Min: 31.73 / Avg: 31.89 / Max: 32.02 Min: 34.26 / Avg: 34.9 / Max: 41.65 Min: 34.27 / Avg: 34.28 / Max: 34.3 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
FS-Mark FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 12.29, N = 15 SE +/- 1.20, N = 3 SE +/- 4.83, N = 3 SE +/- 0.25, N = 3 SE +/- 10.58, N = 15 SE +/- 15.55, N = 3 SE +/- 0.19, N = 3 SE +/- 11.80, N = 15 945 903 984 994 965 970 988 943 1. (CC) gcc options: -static Result Confidence
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 887.4 / Avg: 945.45 / Max: 989.1 Min: 901 / Avg: 902.87 / Max: 905.1 Min: 975.1 / Avg: 983.97 / Max: 991.7 Min: 993.4 / Avg: 993.9 / Max: 994.2 Min: 923.9 / Avg: 965.49 / Max: 1010.6 Min: 938.9 / Avg: 970 / Max: 985.7 Min: 987.6 / Avg: 987.97 / Max: 988.2 Min: 898.5 / Avg: 942.81 / Max: 991.8 1. (CC) gcc options: -static
Timed HMMer Search This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search GRAPHITE LTO LTO GRAPHITE 0.837 1.674 2.511 3.348 4.185 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 3.66 3.66 3.72 3.70 3.49 3.55 3.40 3.38 1. (CC) gcc options: -march=native -pipe -O3 -pthread -lhmmer -lsquid -lm Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.62 / Avg: 3.66 / Max: 3.71 Min: 3.63 / Avg: 3.66 / Max: 3.73 Min: 3.68 / Avg: 3.72 / Max: 3.75 Min: 3.67 / Avg: 3.7 / Max: 3.73 Min: 3.45 / Avg: 3.49 / Max: 3.54 Min: 3.51 / Avg: 3.55 / Max: 3.6 Min: 3.36 / Avg: 3.4 / Max: 3.46 Min: 3.34 / Avg: 3.38 / Max: 3.46 1. (CC) gcc options: -march=native -pipe -O3 -pthread -lhmmer -lsquid -lm
FS-Mark FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 11.67, N = 15 SE +/- 0.59, N = 3 SE +/- 2.53, N = 3 SE +/- 2.44, N = 3 SE +/- 11.43, N = 15 SE +/- 11.69, N = 15 SE +/- 0.85, N = 3 SE +/- 4.66, N = 3 947 917 964 994 966 946 1006 978 1. (CC) gcc options: -static Result Confidence
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 891 / Avg: 947.45 / Max: 991.5 Min: 915.5 / Avg: 916.67 / Max: 917.4 Min: 960.3 / Avg: 964.33 / Max: 969 Min: 989.2 / Avg: 994.03 / Max: 997 Min: 901.2 / Avg: 965.69 / Max: 1011.1 Min: 874.4 / Avg: 945.55 / Max: 989.3 Min: 1004.2 / Avg: 1005.53 / Max: 1007.1 Min: 968.5 / Avg: 977.73 / Max: 983.4 1. (CC) gcc options: -static
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 108.40, N = 3 SE +/- 46.57, N = 3 SE +/- 52.27, N = 3 SE +/- 44.86, N = 3 SE +/- 39.38, N = 3 SE +/- 17.16, N = 3 SE +/- 8.47, N = 3 SE +/- 15.36, N = 3 7399 7497 8045 8116 7489 7517 8017 7962 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 1400 2800 4200 5600 7000 Min: 7193.8 / Avg: 7399.3 / Max: 7561.9 Min: 7416 / Avg: 7497.27 / Max: 7577.3 Min: 7985.8 / Avg: 8045.43 / Max: 8149.6 Min: 8055 / Avg: 8116.03 / Max: 8203.5 Min: 7442.7 / Avg: 7488.63 / Max: 7567 Min: 7486.4 / Avg: 7516.77 / Max: 7545.8 Min: 8000.5 / Avg: 8017.43 / Max: 8026.1 Min: 7940.2 / Avg: 7961.93 / Max: 7991.6 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 103.19, N = 3 SE +/- 1777.12, N = 15 SE +/- 200.09, N = 3 SE +/- 1271.52, N = 15 SE +/- 19.43, N = 3 SE +/- 135.97, N = 3 SE +/- 367.36, N = 3 SE +/- 81.84, N = 3 57825 53144 58085 56185 57984 58167 57614 57950 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 57695.02 / Avg: 57824.62 / Max: 58028.5 Min: 43334.74 / Avg: 53144.26 / Max: 58206.62 Min: 57787.81 / Avg: 58085.29 / Max: 58465.84 Min: 43947.56 / Avg: 56184.85 / Max: 58389.17 Min: 57949.69 / Avg: 57983.61 / Max: 58017 Min: 58007.14 / Avg: 58167.03 / Max: 58437.48 Min: 56895.21 / Avg: 57613.71 / Max: 58105.88 Min: 57792.7 / Avg: 57949.92 / Max: 58067.99 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K SE +/- 18776.70, N = 7 SE +/- 15121.50, N = 3 SE +/- 8402.42, N = 3 SE +/- 24802.99, N = 15 SE +/- 14264.61, N = 15 SE +/- 5273.10, N = 3 SE +/- 14761.95, N = 3 SE +/- 23910.80, N = 15 1490269 1444608 1461562 1417787 1435245 1522336 1460025 1396513 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K Min: 1410468.32 / Avg: 1490268.63 / Max: 1552691.43 Min: 1414364.64 / Avg: 1444607.65 / Max: 1459729.15 Min: 1445306.99 / Avg: 1461562.28 / Max: 1473381.3 Min: 1224148.24 / Avg: 1417786.78 / Max: 1556231 Min: 1307790.55 / Avg: 1435244.66 / Max: 1498171.18 Min: 1512555.39 / Avg: 1522335.89 / Max: 1530642.75 Min: 1443269.91 / Avg: 1460024.76 / Max: 1489454.55 Min: 1203997.65 / Avg: 1396512.51 / Max: 1491624.18 1. (CC) gcc options: -O2
TTSIOD 3D Renderer A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 6.23, N = 3 SE +/- 1.62, N = 3 SE +/- 6.07, N = 3 SE +/- 1.61, N = 3 SE +/- 1.25, N = 3 SE +/- 3.44, N = 3 SE +/- 6.40, N = 3 SE +/- 9.50, N = 3 849 860 856 856 793 803 793 789 1. (CXX) g++ options: -march=native -pipe -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lpthread -fopenmp -fwhole-program -lstdc++ Result Confidence
O2 O3 OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping GRAPHITE LTO LTO GRAPHITE 150 300 450 600 750 Min: 836.54 / Avg: 848.99 / Max: 855.51 Min: 857.05 / Avg: 860.1 / Max: 862.59 Min: 844.31 / Avg: 855.93 / Max: 864.75 Min: 853.17 / Avg: 856.17 / Max: 858.66 Min: 790.45 / Avg: 792.54 / Max: 794.79 Min: 796.69 / Avg: 802.81 / Max: 808.6 Min: 780.03 / Avg: 792.69 / Max: 800.64 Min: 777.85 / Avg: 788.91 / Max: 807.82 1. (CXX) g++ options: -march=native -pipe -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lpthread -fopenmp -fwhole-program -lstdc++
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 1500 3000 4500 6000 7500 SE +/- 61.19, N = 3 SE +/- 11.01, N = 3 SE +/- 38.39, N = 3 SE +/- 49.67, N = 3 SE +/- 42.46, N = 3 SE +/- 34.34, N = 3 SE +/- 26.15, N = 3 SE +/- 48.78, N = 3 6458 6453 7030 6820 6591 6495 7010 6991 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 Min: 6337.1 / Avg: 6458.43 / Max: 6532.9 Min: 6431 / Avg: 6452.6 / Max: 6467.1 Min: 6956.2 / Avg: 7030.43 / Max: 7084.5 Min: 6723.9 / Avg: 6819.53 / Max: 6890.6 Min: 6525.9 / Avg: 6591.03 / Max: 6670.8 Min: 6427.7 / Avg: 6494.6 / Max: 6541.5 Min: 6983.4 / Avg: 7009.7 / Max: 7062 Min: 6936.1 / Avg: 6990.9 / Max: 7088.2 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 SE +/- 1.85, N = 3 SE +/- 3.45, N = 3 SE +/- 2.41, N = 3 SE +/- 2.25, N = 14 SE +/- 4.29, N = 3 SE +/- 2.20, N = 3 SE +/- 4.31, N = 3 SE +/- 4.26, N = 4 246 249 229 244 246 248 237 246 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 242.98 / Avg: 245.54 / Max: 249.13 Min: 245.52 / Avg: 249.05 / Max: 255.94 Min: 225.84 / Avg: 228.97 / Max: 233.71 Min: 229.87 / Avg: 244.21 / Max: 256.13 Min: 238.15 / Avg: 246.16 / Max: 252.82 Min: 243.18 / Avg: 247.51 / Max: 250.38 Min: 231.16 / Avg: 236.78 / Max: 245.24 Min: 233.26 / Avg: 245.94 / Max: 251.67 1. (CC) gcc options: -O2
Apache Siege This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 44.77, N = 3 SE +/- 554.33, N = 4 SE +/- 344.20, N = 3 SE +/- 137.14, N = 3 SE +/- 265.72, N = 3 SE +/- 462.40, N = 9 SE +/- 118.88, N = 3 SE +/- 365.73, N = 3 39625 38571 38871 39720 39041 41310 39402 38006 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 39572.61 / Avg: 39624.98 / Max: 39714.06 Min: 37009.62 / Avg: 38571.15 / Max: 39572.61 Min: 38476.34 / Avg: 38871.18 / Max: 39556.96 Min: 39494.47 / Avg: 39720.26 / Max: 39968.02 Min: 38610.04 / Avg: 39040.7 / Max: 39525.69 Min: 38580.25 / Avg: 41309.55 / Max: 43122.04 Min: 39246.47 / Avg: 39401.82 / Max: 39635.36 Min: 37495.31 / Avg: 38005.72 / Max: 38714.67 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 16.08, N = 3 SE +/- 39.42, N = 3 SE +/- 44.07, N = 3 SE +/- 8.53, N = 3 SE +/- 132.68, N = 3 SE +/- 14.45, N = 3 SE +/- 76.42, N = 3 SE +/- 105.26, N = 3 8928 9234 9500 9471 8867 8999 9467 9596 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 1700 3400 5100 6800 8500 Min: 8897.5 / Avg: 8927.87 / Max: 8952.2 Min: 9164.3 / Avg: 9233.73 / Max: 9300.8 Min: 9426.2 / Avg: 9500.4 / Max: 9578.7 Min: 9454.3 / Avg: 9471.37 / Max: 9479.9 Min: 8699 / Avg: 8867.4 / Max: 9129.2 Min: 8969.9 / Avg: 8998.67 / Max: 9015.4 Min: 9353.1 / Avg: 9466.6 / Max: 9612 Min: 9392.5 / Avg: 9595.6 / Max: 9745.1 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 1.45, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 1.45, N = 3 SE +/- 2.85, N = 3 SE +/- 0.88, N = 3 151 149 150 148 148 151 160 152 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 149 / Avg: 151.33 / Max: 154 Min: 148 / Avg: 149.33 / Max: 151 Min: 148 / Avg: 149.67 / Max: 151 Min: 147 / Avg: 148 / Max: 149 Min: 148 / Avg: 150.67 / Max: 153 Min: 157 / Avg: 160.33 / Max: 166 Min: 151 / Avg: 152.33 / Max: 154 1. (CC) gcc options: -O2 -lpthread
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 17.94, N = 3 SE +/- 21.71, N = 3 SE +/- 26.16, N = 3 SE +/- 35.93, N = 3 SE +/- 12.83, N = 3 SE +/- 23.85, N = 3 SE +/- 16.55, N = 3 SE +/- 87.50, N = 3 9125 9125 9748 9676 9220 9053 9783 9633 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 9089.9 / Avg: 9124.57 / Max: 9149.9 Min: 9081.2 / Avg: 9124.6 / Max: 9147.2 Min: 9706.5 / Avg: 9747.6 / Max: 9796.2 Min: 9604.8 / Avg: 9676.27 / Max: 9718.5 Min: 9196.2 / Avg: 9219.63 / Max: 9240.4 Min: 9004.9 / Avg: 9052.57 / Max: 9077.7 Min: 9764.3 / Avg: 9782.57 / Max: 9815.6 Min: 9501.8 / Avg: 9632.83 / Max: 9798.8 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 SE +/- 5.13, N = 15 SE +/- 6.24, N = 15 SE +/- 4.83, N = 15 SE +/- 4.49, N = 3 SE +/- 1.50, N = 3 SE +/- 5.22, N = 15 SE +/- 5.60, N = 6 SE +/- 3.49, N = 3 414 417 404 401 397 429 402 415 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 Min: 384.85 / Avg: 413.73 / Max: 447.79 Min: 380.51 / Avg: 416.55 / Max: 455.44 Min: 369.81 / Avg: 403.55 / Max: 435.35 Min: 392.21 / Avg: 400.82 / Max: 407.33 Min: 394.23 / Avg: 396.97 / Max: 399.39 Min: 394.48 / Avg: 429.43 / Max: 464.08 Min: 375.31 / Avg: 401.97 / Max: 412.29 Min: 408.66 / Avg: 415.19 / Max: 420.57 1. (CC) gcc options: -O2
Sockperf This is a network socket API performance benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong GRAPHITE LTO LTO GRAPHITE 0.7583 1.5166 2.2749 3.0332 3.7915 SE +/- 0.02, N = 5 SE +/- 0.04, N = 25 SE +/- 0.04, N = 21 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 SE +/- 0.07, N = 25 SE +/- 0.02, N = 5 SE +/- 0.03, N = 25 3.37 3.26 3.31 3.34 3.34 3.12 3.35 3.26 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.3 / Avg: 3.37 / Max: 3.43 Min: 2.57 / Avg: 3.26 / Max: 3.39 Min: 2.56 / Avg: 3.31 / Max: 3.44 Min: 3.31 / Avg: 3.34 / Max: 3.38 Min: 3.27 / Avg: 3.34 / Max: 3.41 Min: 2.54 / Avg: 3.12 / Max: 3.39 Min: 3.29 / Avg: 3.35 / Max: 3.39 Min: 2.58 / Avg: 3.26 / Max: 3.46 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 0.58, N = 3 SE +/- 0.19, N = 3 SE +/- 1.37, N = 3 SE +/- 0.69, N = 3 SE +/- 0.68, N = 3 SE +/- 0.72, N = 3 SE +/- 0.11, N = 3 SE +/- 1.38, N = 3 169 167 157 167 167 166 166 167 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 167.81 / Avg: 168.95 / Max: 169.71 Min: 166.69 / Avg: 167.05 / Max: 167.35 Min: 155.32 / Avg: 157.32 / Max: 159.94 Min: 166.48 / Avg: 167.38 / Max: 168.74 Min: 165.3 / Avg: 166.63 / Max: 167.54 Min: 165.05 / Avg: 166.03 / Max: 167.43 Min: 165.38 / Avg: 165.55 / Max: 165.76 Min: 165.26 / Avg: 167.26 / Max: 169.91 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1000 2000 3000 4000 5000 SE +/- 7.01, N = 3 SE +/- 44.75, N = 15 SE +/- 76.07, N = 4 SE +/- 8.55, N = 3 SE +/- 76.69, N = 4 SE +/- 75.24, N = 3 SE +/- 5.01, N = 3 SE +/- 78.39, N = 4 4477 4417 4495 4329 4418 4658 4366 4558 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 Min: 4467.69 / Avg: 4477.39 / Max: 4490.99 Min: 4222.21 / Avg: 4417.19 / Max: 4595.29 Min: 4267.06 / Avg: 4494.59 / Max: 4583.23 Min: 4311.5 / Avg: 4328.55 / Max: 4338.23 Min: 4320.04 / Avg: 4417.99 / Max: 4646.85 Min: 4507.54 / Avg: 4657.96 / Max: 4736.74 Min: 4356.83 / Avg: 4366.32 / Max: 4373.86 Min: 4328.04 / Avg: 4558.4 / Max: 4679.72 1. (CC) gcc options: -O3
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1500 3000 4500 6000 7500 SE +/- 5.83, N = 3 SE +/- 52.86, N = 3 SE +/- 17.92, N = 3 SE +/- 46.93, N = 3 SE +/- 62.67, N = 3 SE +/- 34.05, N = 3 SE +/- 34.72, N = 3 SE +/- 3.43, N = 3 6771 6498 6659 6421 6422 6859 6906 6594 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 Min: 6759.62 / Avg: 6771.28 / Max: 6777.16 Min: 6392.01 / Avg: 6497.56 / Max: 6555.68 Min: 6623.81 / Avg: 6659.07 / Max: 6682.24 Min: 6371.89 / Avg: 6420.99 / Max: 6514.81 Min: 6322.01 / Avg: 6421.58 / Max: 6537.28 Min: 6822.85 / Avg: 6858.52 / Max: 6926.59 Min: 6855.48 / Avg: 6906.18 / Max: 6972.63 Min: 6587.48 / Avg: 6594.2 / Max: 6598.71 1. (CC) gcc options: -O3
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1000 2000 3000 4000 5000 SE +/- 77.48, N = 3 SE +/- 65.48, N = 4 SE +/- 75.57, N = 3 SE +/- 4.01, N = 3 SE +/- 68.10, N = 4 SE +/- 71.09, N = 3 SE +/- 66.88, N = 4 SE +/- 1.60, N = 3 4350 4311 4222 4187 4430 4500 4364 4458 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 Min: 4228.91 / Avg: 4350.11 / Max: 4494.33 Min: 4128.35 / Avg: 4311.32 / Max: 4408.75 Min: 4135.08 / Avg: 4222.2 / Max: 4372.72 Min: 4180.91 / Avg: 4187.44 / Max: 4194.74 Min: 4226.14 / Avg: 4430.43 / Max: 4500.23 Min: 4358.37 / Avg: 4500.48 / Max: 4575.75 Min: 4163.01 / Avg: 4363.6 / Max: 4434.09 Min: 4455.83 / Avg: 4457.68 / Max: 4460.87 1. (CC) gcc options: -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 237.83, N = 15 SE +/- 431.74, N = 3 SE +/- 303.98, N = 3 SE +/- 210.05, N = 3 SE +/- 167.05, N = 3 SE +/- 238.52, N = 3 SE +/- 155.97, N = 3 SE +/- 210.54, N = 3 23580 22922 24121 24610 24266 23716 23483 23916 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 21527 / Avg: 23580.2 / Max: 24941 Min: 22062 / Avg: 22922 / Max: 23419 Min: 23649 / Avg: 24121.33 / Max: 24689 Min: 24392 / Avg: 24610 / Max: 25030 Min: 24048 / Avg: 24265.67 / Max: 24594 Min: 23241 / Avg: 23716.33 / Max: 23989 Min: 23173 / Avg: 23482.67 / Max: 23670 Min: 23509 / Avg: 23916 / Max: 24213 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Perl Benchmarks Perl benchmark suite that can be used to compare the relative speed of different versions of perl. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter GRAPHITE LTO LTO GRAPHITE 0.0002 0.0004 0.0006 0.0008 0.001 SE +/- 0.00000189, N = 3 SE +/- 0.00000416, N = 3 SE +/- 0.00000106, N = 3 SE +/- 0.00000793, N = 3 SE +/- 0.00000210, N = 3 SE +/- 0.00000487, N = 3 SE +/- 0.00000371, N = 3 SE +/- 0.00000572, N = 3 0.00087047 0.00092120 0.00092415 0.00093417 0.00092345 0.00092592 0.00092216 0.00092737 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter GRAPHITE LTO LTO GRAPHITE 1 2 3 4 5 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0
Sockperf This is a network socket API performance benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 SE +/- 0.20, N = 5 SE +/- 0.43, N = 5 SE +/- 0.22, N = 5 SE +/- 0.85, N = 20 SE +/- 0.22, N = 5 SE +/- 1.22, N = 20 SE +/- 0.30, N = 5 SE +/- 0.37, N = 7 28.21 28.74 28.51 27.14 28.13 26.89 27.90 28.85 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 Min: 27.56 / Avg: 28.21 / Max: 28.61 Min: 27.65 / Avg: 28.74 / Max: 29.97 Min: 28.03 / Avg: 28.51 / Max: 29.32 Min: 11.11 / Avg: 27.14 / Max: 29.47 Min: 27.29 / Avg: 28.13 / Max: 28.6 Min: 3.81 / Avg: 26.89 / Max: 29.73 Min: 26.97 / Avg: 27.9 / Max: 28.62 Min: 27.58 / Avg: 28.85 / Max: 30.04 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.30, N = 15 SE +/- 0.26, N = 15 SE +/- 0.11, N = 3 11.18 11.28 11.23 11.40 11.13 11.94 11.70 11.47 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 11.1 / Avg: 11.18 / Max: 11.26 Min: 11.22 / Avg: 11.28 / Max: 11.33 Min: 11.14 / Avg: 11.23 / Max: 11.32 Min: 11.36 / Avg: 11.4 / Max: 11.44 Min: 11.07 / Avg: 11.13 / Max: 11.22 Min: 11.13 / Avg: 11.94 / Max: 14.38 Min: 11.09 / Avg: 11.7 / Max: 14 Min: 11.34 / Avg: 11.47 / Max: 11.69 1. (CC) gcc options: -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 SE +/- 1.60, N = 3 SE +/- 1.74, N = 3 SE +/- 3.76, N = 4 SE +/- 1.02, N = 3 SE +/- 3.46, N = 3 SE +/- 2.41, N = 15 SE +/- 4.54, N = 3 SE +/- 2.86, N = 3 261 266 248 256 263 263 263 257 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 258.13 / Avg: 260.5 / Max: 263.54 Min: 263.32 / Avg: 265.81 / Max: 269.16 Min: 240.82 / Avg: 248.01 / Max: 257.92 Min: 254.45 / Avg: 256.39 / Max: 257.92 Min: 255.95 / Avg: 262.85 / Max: 266.64 Min: 244.98 / Avg: 263.13 / Max: 277.46 Min: 255.62 / Avg: 262.81 / Max: 271.21 Min: 252.2 / Avg: 256.66 / Max: 261.98 1. (CC) gcc options: -O2
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.48, N = 15 SE +/- 0.88, N = 3 SE +/- 1.66, N = 7 SE +/- 2.10, N = 4 SE +/- 1.53, N = 3 SE +/- 0.67, N = 3 SE +/- 1.69, N = 7 SE +/- 1.86, N = 3 128 131 129 130 125 130 130 134 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 121 / Avg: 128.47 / Max: 141 Min: 129 / Avg: 130.67 / Max: 132 Min: 124 / Avg: 129 / Max: 136 Min: 126 / Avg: 130.25 / Max: 136 Min: 122 / Avg: 125 / Max: 127 Min: 129 / Avg: 129.67 / Max: 131 Min: 124 / Avg: 130.43 / Max: 138 Min: 132 / Avg: 134.33 / Max: 138 1. (CC) gcc options: -O2 -lpthread
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 512 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 1.10, N = 3 SE +/- 2.68, N = 3 SE +/- 1.18, N = 3 SE +/- 1.64, N = 3 SE +/- 1.83, N = 3 SE +/- 3.15, N = 3 SE +/- 0.66, N = 3 SE +/- 2.36, N = 3 9244 8862 9299 9324 8730 9344 8746 8772 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 512 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9242.4 / Avg: 9244.19 / Max: 9246.21 Min: 8856.23 / Avg: 8861.58 / Max: 8864.54 Min: 9297.51 / Avg: 9299.48 / Max: 9301.59 Min: 9320.6 / Avg: 9323.54 / Max: 9326.27 Min: 8726.3 / Avg: 8729.67 / Max: 8732.6 Min: 9339.46 / Avg: 9343.54 / Max: 9349.72 Min: 8744.41 / Avg: 8745.62 / Max: 8746.68 Min: 8767.94 / Avg: 8772.28 / Max: 8776.08 1. (CC) gcc options: -O3 -march=native -pipe
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 128 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 5.45, N = 3 SE +/- 4.35, N = 3 SE +/- 6.02, N = 3 SE +/- 159.15, N = 3 SE +/- 7.97, N = 3 SE +/- 7.69, N = 3 SE +/- 7.10, N = 3 SE +/- 7.66, N = 3 9503 9102 9571 9444 9004 9604 9034 9016 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 128 MiB GRAPHITE LTO LTO GRAPHITE 1700 3400 5100 6800 8500 Min: 9497.68 / Avg: 9503.43 / Max: 9514.33 Min: 9095.15 / Avg: 9101.84 / Max: 9110 Min: 9560.67 / Avg: 9571.03 / Max: 9581.51 Min: 9125.49 / Avg: 9443.6 / Max: 9612.29 Min: 8989 / Avg: 9004.13 / Max: 9016.03 Min: 9592.02 / Avg: 9603.68 / Max: 9618.2 Min: 9022.5 / Avg: 9033.53 / Max: 9046.79 Min: 9006.32 / Avg: 9016.44 / Max: 9031.47 1. (CC) gcc options: -O3 -march=native -pipe
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 34.09, N = 3 SE +/- 49.25, N = 3 SE +/- 25.33, N = 3 SE +/- 61.05, N = 3 SE +/- 76.82, N = 3 SE +/- 18.04, N = 3 SE +/- 16.20, N = 3 SE +/- 82.87, N = 3 9358 9266 9787 9837 9294 9224 9825 9649 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 9292.8 / Avg: 9357.5 / Max: 9408.5 Min: 9176.4 / Avg: 9266.3 / Max: 9346.1 Min: 9754.2 / Avg: 9786.57 / Max: 9836.5 Min: 9737.2 / Avg: 9837.3 / Max: 9947.9 Min: 9140.4 / Avg: 9293.9 / Max: 9376.4 Min: 9189 / Avg: 9223.57 / Max: 9249.8 Min: 9804.9 / Avg: 9824.7 / Max: 9856.8 Min: 9537.9 / Avg: 9648.5 / Max: 9810.7 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 130 260 390 520 650 SE +/- 29.31, N = 3 SE +/- 8.66, N = 3 SE +/- 16.56, N = 3 SE +/- 2.71, N = 3 SE +/- 2.00, N = 3 SE +/- 0.82, N = 3 SE +/- 2.76, N = 3 SE +/- 8.27, N = 3 558 568 577 560 581 595 585 576 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 110 220 330 440 550 Min: 499.16 / Avg: 557.54 / Max: 591.29 Min: 553.3 / Avg: 567.65 / Max: 583.23 Min: 543.79 / Avg: 576.86 / Max: 594.9 Min: 555.17 / Avg: 560.38 / Max: 564.3 Min: 577.86 / Avg: 580.9 / Max: 584.68 Min: 593.85 / Avg: 594.71 / Max: 596.34 Min: 579.78 / Avg: 585.28 / Max: 588.39 Min: 559.78 / Avg: 576.31 / Max: 585.14 1. (CC) gcc options: -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K SE +/- 3050.72, N = 3 SE +/- 2809.10, N = 15 SE +/- 3051.99, N = 3 SE +/- 2904.43, N = 3 SE +/- 2631.54, N = 15 SE +/- 2678.68, N = 13 SE +/- 3972.80, N = 15 SE +/- 3001.51, N = 15 201721 194583 204239 203526 199463 203199 207434 200288 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K Min: 196018.38 / Avg: 201720.66 / Max: 206451.61 Min: 175222.45 / Avg: 194582.87 / Max: 203984.06 Min: 198449.61 / Avg: 204238.74 / Max: 208809.14 Min: 199066.87 / Avg: 203526.04 / Max: 208979.59 Min: 175945.02 / Avg: 199462.58 / Max: 212096.11 Min: 174744.03 / Avg: 203198.57 / Max: 213511.26 Min: 171466.85 / Avg: 207433.51 / Max: 231464.74 Min: 173559.32 / Avg: 200287.76 / Max: 211570.25 1. (CC) gcc options: -O2
Sunflow Rendering System This test runs benchmarks of the Sunflow Rendering System. The Sunflow Rendering System is an open-source render engine for photo-realistic image synthesis with a ray-tracing core. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis GRAPHITE LTO LTO GRAPHITE 0.1823 0.3646 0.5469 0.7292 0.9115 SE +/- 0.01, N = 15 SE +/- 0.01, N = 15 SE +/- 0.01, N = 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 15 SE +/- 0.01, N = 15 SE +/- 0.01, N = 15 SE +/- 0.00, N = 3 0.80 0.80 0.78 0.76 0.81 0.78 0.80 0.77 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 0.73 / Avg: 0.8 / Max: 0.88 Min: 0.74 / Avg: 0.8 / Max: 0.85 Min: 0.71 / Avg: 0.78 / Max: 0.84 Min: 0.74 / Avg: 0.76 / Max: 0.78 Min: 0.75 / Avg: 0.81 / Max: 0.9 Min: 0.72 / Avg: 0.78 / Max: 0.84 Min: 0.74 / Avg: 0.8 / Max: 0.89 Min: 0.77 / Avg: 0.77 / Max: 0.78
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 2.62, N = 3 SE +/- 1.93, N = 3 SE +/- 2.87, N = 3 SE +/- 32.49, N = 3 SE +/- 1.58, N = 3 SE +/- 2.74, N = 3 SE +/- 155.55, N = 3 SE +/- 0.17, N = 3 9128 8723 9168 9195 8655 9216 8776 8671 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9122.61 / Avg: 9127.5 / Max: 9131.59 Min: 8719.38 / Avg: 8723.23 / Max: 8725.26 Min: 9163.86 / Avg: 9168 / Max: 9173.52 Min: 9130.18 / Avg: 9194.97 / Max: 9231.58 Min: 8651.96 / Avg: 8655.11 / Max: 8656.85 Min: 9211.4 / Avg: 9216.45 / Max: 9220.82 Min: 8612.57 / Avg: 8776.32 / Max: 9087.27 Min: 8670.53 / Avg: 8670.87 / Max: 8671.06 1. (CC) gcc options: -O3 -march=native -pipe
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 1.92, N = 3 SE +/- 1.43, N = 3 SE +/- 3.12, N = 3 SE +/- 0.47, N = 3 SE +/- 1.86, N = 3 SE +/- 2.43, N = 3 SE +/- 0.55, N = 3 SE +/- 0.46, N = 3 9097 8699 9145 9093 8629 9181 9104 8648 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9092.74 / Avg: 9096.56 / Max: 9098.85 Min: 8696.98 / Avg: 8699.24 / Max: 8701.88 Min: 9140.87 / Avg: 9144.86 / Max: 9151.01 Min: 9091.73 / Avg: 9092.58 / Max: 9093.35 Min: 8626.61 / Avg: 8628.94 / Max: 8632.62 Min: 9178.13 / Avg: 9181.47 / Max: 9186.19 Min: 9102.89 / Avg: 9103.97 / Max: 9104.68 Min: 8646.87 / Avg: 8647.7 / Max: 8648.47 1. (CC) gcc options: -O3 -march=native -pipe
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 1024 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 2.29, N = 3 SE +/- 0.72, N = 3 SE +/- 0.64, N = 3 SE +/- 0.59, N = 3 SE +/- 1.60, N = 3 SE +/- 2.54, N = 3 SE +/- 2.87, N = 3 SE +/- 0.86, N = 3 9125 8743 9215 9241 8689 9207 8713 8686 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 1024 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9120.6 / Avg: 9125.09 / Max: 9128.13 Min: 8741.27 / Avg: 8742.7 / Max: 8743.49 Min: 9214.08 / Avg: 9214.72 / Max: 9216 Min: 9239.76 / Avg: 9240.77 / Max: 9241.82 Min: 8686.28 / Avg: 8689.47 / Max: 8691.28 Min: 9203.83 / Avg: 9207.17 / Max: 9212.16 Min: 8706.98 / Avg: 8712.68 / Max: 8716.1 Min: 8684.74 / Avg: 8686.32 / Max: 8687.69 1. (CC) gcc options: -O3 -march=native -pipe
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall GRAPHITE LTO LTO GRAPHITE 0.8753 1.7506 2.6259 3.5012 4.3765 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.76 3.66 3.82 3.80 3.66 3.74 3.89 3.83 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.75 / Avg: 3.76 / Max: 3.77 Min: 3.65 / Avg: 3.66 / Max: 3.67 Min: 3.81 / Avg: 3.82 / Max: 3.85 Min: 3.78 / Avg: 3.8 / Max: 3.82 Min: 3.65 / Avg: 3.66 / Max: 3.68 Min: 3.74 / Avg: 3.74 / Max: 3.76 Min: 3.87 / Avg: 3.89 / Max: 3.91 Min: 3.82 / Avg: 3.83 / Max: 3.84 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 40.20, N = 3 SE +/- 172.91, N = 3 SE +/- 39.90, N = 3 SE +/- 139.23, N = 3 SE +/- 149.25, N = 3 SE +/- 162.23, N = 3 SE +/- 93.73, N = 3 SE +/- 75.04, N = 3 18192 17749 18387 17355 17317 18067 18228 18146 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 18130.31 / Avg: 18191.9 / Max: 18267.45 Min: 17412.6 / Avg: 17749.13 / Max: 17986.37 Min: 18326.29 / Avg: 18387.46 / Max: 18462.43 Min: 17083.75 / Avg: 17355.04 / Max: 17545.06 Min: 17163.93 / Avg: 17316.62 / Max: 17615.08 Min: 17827.3 / Avg: 18067.11 / Max: 18376.28 Min: 18061.24 / Avg: 18228.09 / Max: 18385.52 Min: 18063.79 / Avg: 18146.35 / Max: 18296.17 1. (CC) gcc options: -O2
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 1.24, N = 3 SE +/- 112.86, N = 3 SE +/- 4.03, N = 3 SE +/- 170.92, N = 3 SE +/- 2.77, N = 3 SE +/- 2.68, N = 3 SE +/- 2.07, N = 3 SE +/- 1.68, N = 3 9127 9010 8692 8891 8780 9223 9132 8789 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9124.54 / Avg: 9126.98 / Max: 9128.57 Min: 8895.59 / Avg: 9009.87 / Max: 9235.58 Min: 8685.37 / Avg: 8691.7 / Max: 8699.17 Min: 8708.42 / Avg: 8890.52 / Max: 9232.11 Min: 8776.27 / Avg: 8780.07 / Max: 8785.47 Min: 9219.8 / Avg: 9222.57 / Max: 9227.94 Min: 9127.77 / Avg: 9131.9 / Max: 9134.3 Min: 8785.31 / Avg: 8788.53 / Max: 8791 1. (CC) gcc options: -O3 -march=native -pipe
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 0.19, N = 3 SE +/- 2.09, N = 3 SE +/- 1.17, N = 3 SE +/- 1.11, N = 3 SE +/- 5.81, N = 3 SE +/- 2.81, N = 3 SE +/- 5.01, N = 3 SE +/- 3.12, N = 3 9163 8937 8727 8730 8808 9257 9164 8815 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9162.98 / Avg: 9163.29 / Max: 9163.64 Min: 8934.58 / Avg: 8937.18 / Max: 8941.32 Min: 8724.51 / Avg: 8726.83 / Max: 8728.18 Min: 8729.19 / Avg: 8730.43 / Max: 8732.65 Min: 8796.55 / Avg: 8807.96 / Max: 8815.58 Min: 9252.01 / Avg: 9257.18 / Max: 9261.68 Min: 9154.43 / Avg: 9164.41 / Max: 9170.21 Min: 8809 / Avg: 8814.72 / Max: 8819.74 1. (CC) gcc options: -O3 -march=native -pipe
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 6.48, N = 3 SE +/- 2.53, N = 3 SE +/- 0.45, N = 3 SE +/- 9.28, N = 3 SE +/- 2.01, N = 3 SE +/- 6.31, N = 3 SE +/- 4.53, N = 3 SE +/- 2.63, N = 3 1205 1226 1278 1226 1242 1229 1228 1225 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1191.89 / Avg: 1204.74 / Max: 1212.59 Min: 1220.91 / Avg: 1225.89 / Max: 1229.1 Min: 1277.42 / Avg: 1278.3 / Max: 1278.92 Min: 1207.59 / Avg: 1225.89 / Max: 1237.78 Min: 1238.62 / Avg: 1241.8 / Max: 1245.52 Min: 1222.39 / Avg: 1228.94 / Max: 1241.55 Min: 1219.38 / Avg: 1228.16 / Max: 1234.48 Min: 1220.86 / Avg: 1225.24 / Max: 1229.94 1. (CC) gcc options: -O3
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 1.63, N = 3 SE +/- 0.48, N = 3 SE +/- 3.40, N = 3 SE +/- 3.01, N = 3 SE +/- 1.90, N = 3 SE +/- 2.67, N = 3 SE +/- 2.21, N = 3 SE +/- 2.10, N = 3 9152 8901 9186 8796 8875 9328 9230 8884 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9149.3 / Avg: 9152.09 / Max: 9154.95 Min: 8900.1 / Avg: 8900.72 / Max: 8901.65 Min: 9179.65 / Avg: 9186.34 / Max: 9190.72 Min: 8790.32 / Avg: 8795.75 / Max: 8800.71 Min: 8871.87 / Avg: 8875.41 / Max: 8878.38 Min: 9322.83 / Avg: 9327.88 / Max: 9331.93 Min: 9226.87 / Avg: 9229.77 / Max: 9234.11 Min: 8880.25 / Avg: 8884.14 / Max: 8887.45 1. (CC) gcc options: -O3 -march=native -pipe
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 5.39, N = 3 SE +/- 3.01, N = 3 SE +/- 0.63, N = 3 SE +/- 3.11, N = 3 SE +/- 0.85, N = 3 SE +/- 5.09, N = 3 SE +/- 2.38, N = 3 SE +/- 4.85, N = 3 1213 1233 1282 1233 1247 1261 1230 1209 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1204 / Avg: 1213.21 / Max: 1222.67 Min: 1227.08 / Avg: 1233.02 / Max: 1236.83 Min: 1281.06 / Avg: 1282.15 / Max: 1283.23 Min: 1227.05 / Avg: 1233.26 / Max: 1236.68 Min: 1245.57 / Avg: 1246.82 / Max: 1248.44 Min: 1251.26 / Avg: 1261.41 / Max: 1267.13 Min: 1226.06 / Avg: 1230.13 / Max: 1234.31 Min: 1199.16 / Avg: 1208.73 / Max: 1214.84 1. (CC) gcc options: -O3
Tinymembench This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 68.34, N = 3 SE +/- 37.35, N = 3 SE +/- 20.59, N = 3 SE +/- 55.09, N = 3 SE +/- 29.34, N = 3 SE +/- 106.07, N = 3 SE +/- 30.81, N = 3 SE +/- 63.95, N = 3 9556 9334 9586 9259 9176 9712 9159 9386 1. (CC) gcc options: -O2 -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 9424.2 / Avg: 9556.37 / Max: 9652.6 Min: 9271.5 / Avg: 9333.57 / Max: 9400.6 Min: 9564.5 / Avg: 9585.93 / Max: 9627.1 Min: 9149 / Avg: 9258.93 / Max: 9320.2 Min: 9120.4 / Avg: 9175.77 / Max: 9220.3 Min: 9574.6 / Avg: 9712.3 / Max: 9920.9 Min: 9103.8 / Avg: 9159.47 / Max: 9210.2 Min: 9281.2 / Avg: 9386.03 / Max: 9501.9 1. (CC) gcc options: -O2 -march=native -pipe -O3 -lm
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 SE +/- 99.36, N = 3 SE +/- 2.65, N = 3 SE +/- 8.97, N = 3 SE +/- 12.53, N = 3 SE +/- 8.52, N = 3 SE +/- 1.50, N = 3 SE +/- 3.89, N = 3 SE +/- 14.47, N = 3 6961 6907 6945 7174 6861 7275 7246 6889 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 Min: 6855.31 / Avg: 6961.19 / Max: 7159.75 Min: 6901.71 / Avg: 6906.83 / Max: 6910.56 Min: 6926.94 / Avg: 6944.88 / Max: 6954.03 Min: 7153.44 / Avg: 7173.59 / Max: 7196.56 Min: 6851.56 / Avg: 6860.68 / Max: 6877.69 Min: 7273.86 / Avg: 7275.48 / Max: 7278.48 Min: 7238.86 / Avg: 7246.34 / Max: 7251.91 Min: 6862.77 / Avg: 6889.22 / Max: 6912.61 1. (CC) gcc options: -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 111.95, N = 15 SE +/- 145.97, N = 15 SE +/- 140.08, N = 5 SE +/- 45.61, N = 3 SE +/- 140.95, N = 3 SE +/- 14.02, N = 3 SE +/- 135.31, N = 3 SE +/- 136.36, N = 5 9257 9025 9150 8931 9468 9127 9242 9088 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 7905.8 / Avg: 9256.99 / Max: 9590.6 Min: 7842.1 / Avg: 9025.19 / Max: 9504.7 Min: 8861.2 / Avg: 9149.66 / Max: 9510.4 Min: 8867.9 / Avg: 8931 / Max: 9019.6 Min: 9231.3 / Avg: 9468.33 / Max: 9719 Min: 9098.7 / Avg: 9126.63 / Max: 9142.8 Min: 9002.4 / Avg: 9241.77 / Max: 9470.8 Min: 8894.6 / Avg: 9088.36 / Max: 9620.9 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 70.15, N = 3 SE +/- 67.00, N = 3 SE +/- 143.99, N = 4 SE +/- 42.52, N = 3 SE +/- 61.61, N = 3 SE +/- 73.66, N = 3 SE +/- 33.56, N = 3 SE +/- 68.23, N = 3 9087 9048 8858 8660 8716 9128 9180 9118 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 8956.06 / Avg: 9087.05 / Max: 9196.06 Min: 8919.24 / Avg: 9047.79 / Max: 9144.82 Min: 8621.85 / Avg: 8858.13 / Max: 9253.9 Min: 8602.15 / Avg: 8660.38 / Max: 8743.17 Min: 8602.73 / Avg: 8715.67 / Max: 8814.82 Min: 8981.2 / Avg: 9127.76 / Max: 9213.94 Min: 9140.9 / Avg: 9180.48 / Max: 9247.22 Min: 9026.8 / Avg: 9117.61 / Max: 9251.23 1. (CC) gcc options: -O2
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 3.52, N = 3 SE +/- 3.46, N = 3 SE +/- 2.66, N = 3 SE +/- 2.95, N = 3 SE +/- 3.31, N = 3 SE +/- 5.34, N = 3 SE +/- 5.00, N = 3 SE +/- 1.76, N = 3 9055 8829 9116 8629 8735 9147 9064 8720 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9048.48 / Avg: 9055.31 / Max: 9060.16 Min: 8822.46 / Avg: 8829.36 / Max: 8833.3 Min: 9112.19 / Avg: 9116.42 / Max: 9121.32 Min: 8623.44 / Avg: 8629.2 / Max: 8633.21 Min: 8728.83 / Avg: 8735.37 / Max: 8739.56 Min: 9136.09 / Avg: 9146.75 / Max: 9152.75 Min: 9053.98 / Avg: 9063.97 / Max: 9069.49 Min: 8716.88 / Avg: 8720.19 / Max: 8722.86 1. (CC) gcc options: -O3 -march=native -pipe
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.A GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 187.72, N = 3 SE +/- 68.38, N = 3 SE +/- 225.23, N = 3 SE +/- 89.77, N = 3 SE +/- 72.47, N = 3 SE +/- 213.04, N = 3 SE +/- 404.10, N = 3 SE +/- 189.37, N = 3 48667 47581 48142 47016 49301 47581 48045 46536 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.A GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 48307.19 / Avg: 48666.83 / Max: 48940 Min: 47444.97 / Avg: 47581.29 / Max: 47659 Min: 47769.48 / Avg: 48142.12 / Max: 48547.63 Min: 46838.43 / Avg: 47016.21 / Max: 47126.88 Min: 49156.31 / Avg: 49301.04 / Max: 49380.15 Min: 47185.53 / Avg: 47581.18 / Max: 47915.95 Min: 47579.34 / Avg: 48044.81 / Max: 48849.73 Min: 46260.16 / Avg: 46536.17 / Max: 46898.78 1. 3.3
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Thread GRAPHITE LTO LTO GRAPHITE 0.567 1.134 1.701 2.268 2.835 SE +/- 0.04, N = 4 SE +/- 0.03, N = 7 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 10 SE +/- 0.04, N = 4 2.46 2.44 2.44 2.43 2.38 2.52 2.44 2.44 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Thread GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.38 / Avg: 2.46 / Max: 2.54 Min: 2.36 / Avg: 2.44 / Max: 2.56 Min: 2.4 / Avg: 2.44 / Max: 2.51 Min: 2.4 / Avg: 2.43 / Max: 2.48 Min: 2.37 / Avg: 2.38 / Max: 2.4 Min: 2.42 / Avg: 2.52 / Max: 2.57 Min: 2.35 / Avg: 2.44 / Max: 2.57 Min: 2.39 / Avg: 2.44 / Max: 2.56 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K SE +/- 1193.86, N = 3 SE +/- 742.31, N = 3 SE +/- 325.29, N = 3 SE +/- 338.21, N = 3 SE +/- 396.38, N = 3 SE +/- 359.99, N = 3 SE +/- 407.46, N = 3 SE +/- 642.35, N = 3 70199 68632 72618 70169 72275 71597 72665 71510 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K Min: 67931.54 / Avg: 70198.83 / Max: 71980.88 Min: 67222.48 / Avg: 68632 / Max: 69740.52 Min: 72082.22 / Avg: 72618.44 / Max: 73205.61 Min: 69508.55 / Avg: 70169.09 / Max: 70625.56 Min: 71784.09 / Avg: 72274.78 / Max: 73059.36 Min: 70958.35 / Avg: 71597 / Max: 72204.2 Min: 72224.57 / Avg: 72664.73 / Max: 73478.76 Min: 70445.79 / Avg: 71509.91 / Max: 72665.34 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1500 3000 4500 6000 7500 SE +/- 20.61, N = 3 SE +/- 9.55, N = 3 SE +/- 8.93, N = 3 SE +/- 23.79, N = 3 SE +/- 13.25, N = 3 SE +/- 13.19, N = 3 SE +/- 15.29, N = 3 SE +/- 18.69, N = 3 7029 6685 6750 6972 6683 7074 7009 6682 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 Min: 6988.82 / Avg: 7028.91 / Max: 7057.23 Min: 6666.01 / Avg: 6685.02 / Max: 6696.14 Min: 6732.62 / Avg: 6750.4 / Max: 6760.84 Min: 6924.76 / Avg: 6972.18 / Max: 6999.2 Min: 6656.71 / Avg: 6683.07 / Max: 6698.54 Min: 7048.31 / Avg: 7074.34 / Max: 7091.06 Min: 6978.54 / Avg: 7008.56 / Max: 7028.68 Min: 6648.08 / Avg: 6682.17 / Max: 6712.48 1. (CC) gcc options: -O3
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 SE +/- 0.31, N = 3 SE +/- 0.25, N = 3 SE +/- 0.25, N = 3 SE +/- 0.31, N = 6 SE +/- 0.31, N = 6 SE +/- 0.12, N = 3 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 22.10 23.03 22.70 23.39 22.97 22.62 22.67 22.81 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 Min: 21.51 / Avg: 22.1 / Max: 22.53 Min: 22.6 / Avg: 23.03 / Max: 23.45 Min: 22.21 / Avg: 22.7 / Max: 23.02 Min: 22.86 / Avg: 23.39 / Max: 24.81 Min: 22.27 / Avg: 22.97 / Max: 24.16 Min: 22.38 / Avg: 22.62 / Max: 22.78 Min: 22.4 / Avg: 22.67 / Max: 23.02 Min: 22.69 / Avg: 22.81 / Max: 23.06 1. (CC) gcc options: -march=native -pipe -O3 -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 2.00, N = 3 SE +/- 2.33, N = 3 SE +/- 1.00, N = 3 SE +/- 1.76, N = 3 SE +/- 2.08, N = 3 SE +/- 2.19, N = 3 SE +/- 2.08, N = 3 125 123 126 121 124 127 120 123 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 121 / Avg: 125 / Max: 127 Min: 124 / Avg: 126.33 / Max: 131 Min: 120 / Avg: 121 / Max: 123 Min: 121 / Avg: 124.33 / Max: 127 Min: 124 / Avg: 127 / Max: 131 Min: 116 / Avg: 120.33 / Max: 123 Min: 119 / Avg: 123 / Max: 126 1. (CC) gcc options: -O2 -lpthread
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 SE +/- 0.28, N = 3 SE +/- 0.47, N = 4 SE +/- 0.19, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.52, N = 3 SE +/- 0.16, N = 3 SE +/- 0.32, N = 3 26.15 27.65 26.42 27.25 26.28 27.30 26.52 26.99 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 Min: 25.72 / Avg: 26.15 / Max: 26.68 Min: 26.98 / Avg: 27.65 / Max: 29.03 Min: 26.03 / Avg: 26.42 / Max: 26.62 Min: 27.21 / Avg: 27.25 / Max: 27.29 Min: 26.12 / Avg: 26.28 / Max: 26.42 Min: 26.29 / Avg: 27.3 / Max: 28 Min: 26.26 / Avg: 26.52 / Max: 26.8 Min: 26.55 / Avg: 26.99 / Max: 27.6 1. (CC) gcc options: -march=native -pipe -O3 -lm
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 70.86, N = 3 SE +/- 68.39, N = 3 SE +/- 182.26, N = 3 SE +/- 107.20, N = 11 SE +/- 24.58, N = 3 SE +/- 30.69, N = 3 SE +/- 105.30, N = 15 SE +/- 65.70, N = 3 10356 10199 10783 10635 10424 10264 10493 10719 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 10215 / Avg: 10356.33 / Max: 10436 Min: 10121 / Avg: 10198.67 / Max: 10335 Min: 10419 / Avg: 10783 / Max: 10982 Min: 9704.5 / Avg: 10635.14 / Max: 10900 Min: 10375 / Avg: 10424 / Max: 10452 Min: 10213 / Avg: 10263.67 / Max: 10319 Min: 9679.9 / Avg: 10492.95 / Max: 10850 Min: 10642 / Avg: 10719.33 / Max: 10850 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
t-test1 This is a test of t-test1 for basic memory allocator benchmarks. Note this test profile is currently very basic and the overall time does include the warmup time of the custom t-test1 compilation. Improvements welcome. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.18, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.16, N = 3 15.52 15.43 16.30 15.78 15.47 15.44 15.42 15.73 1. (CC) gcc options: -pthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 15.41 / Avg: 15.52 / Max: 15.67 Min: 15.4 / Avg: 15.43 / Max: 15.49 Min: 15.96 / Avg: 16.3 / Max: 16.55 Min: 15.76 / Avg: 15.78 / Max: 15.8 Min: 15.36 / Avg: 15.47 / Max: 15.64 Min: 15.35 / Avg: 15.44 / Max: 15.57 Min: 15.41 / Avg: 15.42 / Max: 15.44 Min: 15.41 / Avg: 15.73 / Max: 15.91 1. (CC) gcc options: -pthread -march=native -pipe -O3
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 8.29, N = 4 SE +/- 33.59, N = 4 SE +/- 58.55, N = 4 SE +/- 24.86, N = 4 SE +/- 28.78, N = 4 SE +/- 50.27, N = 4 SE +/- 48.42, N = 4 SE +/- 48.55, N = 4 3353 3441 3459 3334 3340 3410 3398 3524 Result Confidence
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3328 / Avg: 3352.5 / Max: 3364 Min: 3380 / Avg: 3440.75 / Max: 3536 Min: 3350 / Avg: 3459.25 / Max: 3561 Min: 3270 / Avg: 3334.25 / Max: 3390 Min: 3267 / Avg: 3339.5 / Max: 3395 Min: 3280 / Avg: 3410 / Max: 3521 Min: 3296 / Avg: 3397.75 / Max: 3500 Min: 3406 / Avg: 3523.75 / Max: 3613
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.00, N = 3 SE +/- 1.70, N = 15 SE +/- 2.25, N = 4 SE +/- 1.53, N = 3 SE +/- 1.71, N = 6 SE +/- 1.45, N = 15 127 123 128 128 130 124 128 127 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 122 / Avg: 123 / Max: 125 Min: 120 / Avg: 127.93 / Max: 145 Min: 123 / Avg: 129.5 / Max: 133 Min: 121 / Avg: 124 / Max: 126 Min: 123 / Avg: 128 / Max: 133 Min: 119 / Avg: 127 / Max: 137 1. (CC) gcc options: -O2 -lpthread
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1500 3000 4500 6000 7500 SE +/- 11.20, N = 3 SE +/- 120.56, N = 3 SE +/- 28.33, N = 3 SE +/- 5.13, N = 3 SE +/- 133.98, N = 3 SE +/- 17.09, N = 3 SE +/- 124.99, N = 3 SE +/- 13.56, N = 3 6881 6655 6610 6898 6705 6908 6736 6544 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 Min: 6861.27 / Avg: 6880.67 / Max: 6900.08 Min: 6523.21 / Avg: 6655.04 / Max: 6895.81 Min: 6553.1 / Avg: 6609.57 / Max: 6641.81 Min: 6888.52 / Avg: 6898.46 / Max: 6905.62 Min: 6567.41 / Avg: 6705.1 / Max: 6973.03 Min: 6886.42 / Avg: 6908.39 / Max: 6942.05 Min: 6485.77 / Avg: 6735.68 / Max: 6866.03 Min: 6519.97 / Avg: 6543.58 / Max: 6566.94 1. (CC) gcc options: -O3
R Benchmark This test is a quick-running survey of general R performance Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better R Benchmark GRAPHITE LTO LTO GRAPHITE 0.1058 0.2116 0.3174 0.4232 0.529 SE +/- 0.0011, N = 3 SE +/- 0.0009, N = 3 SE +/- 0.0008, N = 3 SE +/- 0.0003, N = 3 SE +/- 0.0019, N = 3 SE +/- 0.0018, N = 3 SE +/- 0.0005, N = 3 SE +/- 0.0003, N = 3 0.4703 0.4610 0.4611 0.4630 0.4474 0.4516 0.4458 0.4487 1. R scripting front-end version 3.6.0 (2019-04-26) Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better R Benchmark GRAPHITE LTO LTO GRAPHITE 1 2 3 4 5 Min: 0.47 / Avg: 0.47 / Max: 0.47 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.44 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.44 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.45 1. R scripting front-end version 3.6.0 (2019-04-26)
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 900 1800 2700 3600 4500 SE +/- 69.63, N = 3 SE +/- 3.80, N = 3 SE +/- 75.27, N = 3 SE +/- 62.83, N = 3 SE +/- 2.15, N = 3 SE +/- 65.11, N = 3 SE +/- 72.23, N = 3 SE +/- 65.68, N = 3 4316 4316 4188 4384 4417 4409 4221 4246 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 Min: 4176.29 / Avg: 4315.54 / Max: 4386.37 Min: 4308.15 / Avg: 4315.74 / Max: 4319.79 Min: 4100.14 / Avg: 4187.96 / Max: 4337.75 Min: 4258.94 / Avg: 4384.14 / Max: 4455.96 Min: 4414.39 / Avg: 4417.31 / Max: 4421.51 Min: 4278.63 / Avg: 4408.85 / Max: 4474.04 Min: 4144.15 / Avg: 4220.85 / Max: 4365.22 Min: 4178.86 / Avg: 4246.24 / Max: 4377.58 1. (CC) gcc options: -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 0.88, N = 3 SE +/- 1.91, N = 4 SE +/- 0.67, N = 3 SE +/- 1.53, N = 3 SE +/- 1.15, N = 3 SE +/- 1.00, N = 3 111 112 114 116 114 110 111 113 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 109 / Avg: 110.67 / Max: 112 Min: 109 / Avg: 112 / Max: 117 Min: 113 / Avg: 113.67 / Max: 115 Min: 112 / Avg: 114 / Max: 117 Min: 109 / Avg: 111 / Max: 113 Min: 112 / Avg: 113 / Max: 115 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.93, N = 4 SE +/- 2.08, N = 3 SE +/- 1.67, N = 3 SE +/- 1.70, N = 4 SE +/- 1.67, N = 3 SE +/- 1.65, N = 6 113 110 112 114 113 114 116 110 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 111 / Avg: 113.25 / Max: 119 Min: 106 / Avg: 110 / Max: 113 Min: 110 / Avg: 111.67 / Max: 115 Min: 111 / Avg: 114.25 / Max: 119 Min: 111 / Avg: 114.33 / Max: 116 Min: 112 / Avg: 115.5 / Max: 123 1. (CC) gcc options: -O2 -lpthread
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 9.51, N = 3 SE +/- 6.68, N = 3 SE +/- 0.83, N = 3 SE +/- 6.80, N = 3 SE +/- 5.17, N = 3 SE +/- 5.03, N = 3 SE +/- 7.04, N = 3 SE +/- 3.02, N = 3 9021 8792 9082 9011 8683 9144 9041 8697 1. (CC) gcc options: -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 Min: 9006.37 / Avg: 9021.27 / Max: 9038.97 Min: 8780.53 / Avg: 8791.82 / Max: 8803.65 Min: 9080.57 / Avg: 9082.23 / Max: 9083.09 Min: 8997.74 / Avg: 9011.22 / Max: 9019.53 Min: 8674.3 / Avg: 8683.23 / Max: 8692.22 Min: 9135.74 / Avg: 9144.26 / Max: 9153.15 Min: 9027.45 / Avg: 9041.49 / Max: 9049.26 Min: 8693.01 / Avg: 8696.75 / Max: 8702.73 1. (CC) gcc options: -O3 -march=native -pipe
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 5.65, N = 3 SE +/- 5.99, N = 3 SE +/- 5.59, N = 3 SE +/- 4.49, N = 3 SE +/- 17.39, N = 3 SE +/- 10.88, N = 3 SE +/- 4.80, N = 3 SE +/- 2.37, N = 3 1099 1109 1124 1088 1076 1068 1108 1079 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1087.75 / Avg: 1098.9 / Max: 1106.07 Min: 1097.31 / Avg: 1109.1 / Max: 1116.87 Min: 1117.58 / Avg: 1124.39 / Max: 1135.48 Min: 1082.97 / Avg: 1087.82 / Max: 1096.79 Min: 1041.93 / Avg: 1076 / Max: 1099.1 Min: 1052.43 / Avg: 1068.45 / Max: 1089.21 Min: 1101.93 / Avg: 1107.88 / Max: 1117.38 Min: 1076.44 / Avg: 1079.12 / Max: 1083.85 1. (CC) gcc options: -O3
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls GRAPHITE LTO LTO GRAPHITE 0.5445 1.089 1.6335 2.178 2.7225 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 2.35 2.30 2.39 2.38 2.31 2.34 2.42 2.39 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.34 / Avg: 2.35 / Max: 2.35 Min: 2.3 / Avg: 2.3 / Max: 2.3 Min: 2.38 / Avg: 2.39 / Max: 2.39 Min: 2.38 / Avg: 2.38 / Max: 2.38 Min: 2.3 / Avg: 2.31 / Max: 2.32 Min: 2.33 / Avg: 2.34 / Max: 2.35 Min: 2.4 / Avg: 2.42 / Max: 2.43 Min: 2.39 / Avg: 2.39 / Max: 2.4 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 885.20, N = 3 SE +/- 1839.13, N = 15 SE +/- 1292.44, N = 3 SE +/- 1737.85, N = 3 SE +/- 1669.29, N = 15 SE +/- 955.72, N = 3 SE +/- 482.93, N = 3 SE +/- 1600.37, N = 15 103102 100686 104494 102842 101173 104054 105877 105017 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 101910.83 / Avg: 103102.35 / Max: 104832.11 Min: 83170.89 / Avg: 100686.06 / Max: 106400.67 Min: 101910.83 / Avg: 104493.74 / Max: 105872.62 Min: 100946.37 / Avg: 102841.5 / Max: 106312.29 Min: 85676.04 / Avg: 101173.06 / Max: 109028.96 Min: 102976.67 / Avg: 104054.25 / Max: 105960.27 Min: 105090.31 / Avg: 105877.02 / Max: 106755.63 Min: 90395.48 / Avg: 105016.57 / Max: 115004.49 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 6.09, N = 3 SE +/- 3.75, N = 3 SE +/- 2.01, N = 3 SE +/- 9.96, N = 3 SE +/- 4.02, N = 3 SE +/- 3.18, N = 3 SE +/- 3.62, N = 3 SE +/- 8.50, N = 3 1230 1251 1286 1257 1254 1267 1223 1236 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1218.5 / Avg: 1230.34 / Max: 1238.73 Min: 1243.51 / Avg: 1250.98 / Max: 1255.37 Min: 1282.41 / Avg: 1286.44 / Max: 1288.62 Min: 1245.98 / Avg: 1256.93 / Max: 1276.82 Min: 1246.12 / Avg: 1254.14 / Max: 1258.66 Min: 1261.29 / Avg: 1267.28 / Max: 1272.13 Min: 1216.73 / Avg: 1223.47 / Max: 1229.15 Min: 1219.97 / Avg: 1236.31 / Max: 1248.55 1. (CC) gcc options: -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.83, N = 15 SE +/- 1.60, N = 7 SE +/- 1.51, N = 7 SE +/- 1.86, N = 3 SE +/- 1.15, N = 3 SE +/- 1.20, N = 3 SE +/- 2.03, N = 3 117 121 123 119 120 118 117 118 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 111 / Avg: 120.8 / Max: 138 Min: 116 / Avg: 123.29 / Max: 130 Min: 115 / Avg: 119.29 / Max: 125 Min: 116 / Avg: 119.67 / Max: 122 Min: 116 / Avg: 118 / Max: 120 Min: 115 / Avg: 117.33 / Max: 119 Min: 114 / Avg: 117.67 / Max: 121 1. (CC) gcc options: -O2 -lpthread
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.37, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 28.31 28.32 28.20 28.22 26.94 27.18 27.29 27.30 1. (CC) gcc options: -lm -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 Min: 28.31 / Avg: 28.31 / Max: 28.33 Min: 28.32 / Avg: 28.32 / Max: 28.33 Min: 28.19 / Avg: 28.2 / Max: 28.21 Min: 28.19 / Avg: 28.22 / Max: 28.26 Min: 26.82 / Avg: 26.94 / Max: 27.17 Min: 26.8 / Avg: 27.18 / Max: 27.91 Min: 27.28 / Avg: 27.29 / Max: 27.3 Min: 27.27 / Avg: 27.3 / Max: 27.34 1. (CC) gcc options: -lm -O3 -march=native -pipe
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K SE +/- 285.26, N = 3 SE +/- 266.65, N = 3 SE +/- 332.92, N = 3 SE +/- 425.19, N = 3 SE +/- 158.04, N = 3 SE +/- 207.85, N = 3 SE +/- 133.16, N = 3 SE +/- 120.38, N = 3 36324 34922 35875 35605 36699 36650 36185 36061 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 35874.44 / Avg: 36324.25 / Max: 36853.09 Min: 34508.32 / Avg: 34921.88 / Max: 35420.27 Min: 35293.31 / Avg: 35874.76 / Max: 36446.47 Min: 34846.53 / Avg: 35605.28 / Max: 36317.21 Min: 36392.07 / Avg: 36699.49 / Max: 36916.87 Min: 36330.09 / Avg: 36649.69 / Max: 37039.72 Min: 35934.87 / Avg: 36184.73 / Max: 36389.48 Min: 35839.28 / Avg: 36061.38 / Max: 36252.92 1. (CC) gcc options: -O2
Opus Codec Encoding Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus and then to decode the generated Opus file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.0.1 WAV, Opus Encode, Opus Decode GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.01, N = 5 SE +/- 0.00, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 8.32 8.15 8.19 8.19 8.12 8.13 7.92 8.07 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -pie -logg -lopus -lm Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.0.1 WAV, Opus Encode, Opus Decode GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 8.31 / Avg: 8.32 / Max: 8.34 Min: 8.14 / Avg: 8.15 / Max: 8.15 Min: 8.18 / Avg: 8.19 / Max: 8.21 Min: 8.18 / Avg: 8.19 / Max: 8.2 Min: 8.08 / Avg: 8.12 / Max: 8.14 Min: 8.11 / Avg: 8.13 / Max: 8.14 Min: 7.88 / Avg: 7.92 / Max: 7.93 Min: 8.04 / Avg: 8.07 / Max: 8.09 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -pie -logg -lopus -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 160K 320K 480K 640K 800K SE +/- 14248.28, N = 3 SE +/- 9543.92, N = 15 SE +/- 10320.72, N = 15 SE +/- 6191.70, N = 3 SE +/- 12413.45, N = 15 SE +/- 12591.18, N = 4 SE +/- 10287.78, N = 15 SE +/- 10934.32, N = 3 718826 738183 739121 744469 724827 739881 746760 754554 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 130K 260K 390K 520K 650K Min: 693766.94 / Avg: 718826.45 / Max: 743105.95 Min: 645242.6 / Avg: 738182.61 / Max: 785878.74 Min: 649334.18 / Avg: 739120.6 / Max: 775757.58 Min: 737752.16 / Avg: 744468.52 / Max: 756836.66 Min: 628992.63 / Avg: 724826.95 / Max: 778115.5 Min: 702331.96 / Avg: 739881.45 / Max: 754605.75 Min: 662783.17 / Avg: 746759.99 / Max: 803137.26 Min: 734576.76 / Avg: 754553.59 / Max: 772247.36 1. (CC) gcc options: -O2
Sudokut This is a test of Sudokut, which is a Sudoku puzzle solver written in Tcl. This test measures how long it takes to solve 100 Sudoku puzzles. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sudokut 0.4 Total Time GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 9.84 9.80 9.89 9.94 9.72 9.74 9.55 9.47 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sudokut 0.4 Total Time GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 9.83 / Avg: 9.84 / Max: 9.84 Min: 9.79 / Avg: 9.8 / Max: 9.8 Min: 9.88 / Avg: 9.89 / Max: 9.89 Min: 9.94 / Avg: 9.94 / Max: 9.95 Min: 9.71 / Avg: 9.72 / Max: 9.72 Min: 9.73 / Avg: 9.74 / Max: 9.74 Min: 9.54 / Avg: 9.55 / Max: 9.55 Min: 9.47 / Avg: 9.47 / Max: 9.47
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 SE +/- 1.21, N = 3 SE +/- 5.12, N = 3 SE +/- 2.43, N = 3 SE +/- 5.69, N = 3 SE +/- 3.14, N = 3 SE +/- 0.54, N = 3 SE +/- 2.28, N = 3 SE +/- 1.98, N = 3 319 304 305 307 309 307 311 308 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 Min: 317.29 / Avg: 318.5 / Max: 320.92 Min: 298.38 / Avg: 303.79 / Max: 314.03 Min: 300.17 / Avg: 304.82 / Max: 308.38 Min: 295.77 / Avg: 307.15 / Max: 312.97 Min: 302.83 / Avg: 308.77 / Max: 313.51 Min: 306 / Avg: 306.92 / Max: 307.87 Min: 308.02 / Avg: 310.85 / Max: 315.37 Min: 304.51 / Avg: 308.03 / Max: 311.36 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 100 200 300 400 500 SE +/- 3.54, N = 3 SE +/- 8.80, N = 3 SE +/- 2.26, N = 3 SE +/- 5.66, N = 3 SE +/- 3.24, N = 3 SE +/- 3.25, N = 3 SE +/- 6.22, N = 3 SE +/- 2.74, N = 3 452 447 437 454 456 458 442 445 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 Min: 446.09 / Avg: 452.46 / Max: 458.33 Min: 432.39 / Avg: 446.71 / Max: 462.72 Min: 433.4 / Avg: 437.46 / Max: 441.22 Min: 442.53 / Avg: 453.85 / Max: 459.72 Min: 450.66 / Avg: 456.25 / Max: 461.88 Min: 453.71 / Avg: 457.79 / Max: 464.22 Min: 434.96 / Avg: 441.86 / Max: 454.27 Min: 439.37 / Avg: 444.8 / Max: 448.13 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 SE +/- 1.21, N = 3 SE +/- 2.53, N = 3 SE +/- 1.04, N = 3 SE +/- 2.62, N = 3 SE +/- 2.10, N = 3 SE +/- 3.54, N = 4 SE +/- 2.91, N = 3 SE +/- 1.25, N = 3 241 234 230 238 236 235 236 236 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 Min: 239.37 / Avg: 241.04 / Max: 243.39 Min: 228.86 / Avg: 233.87 / Max: 236.93 Min: 228.84 / Avg: 230.48 / Max: 232.42 Min: 233.54 / Avg: 238.03 / Max: 242.61 Min: 231.5 / Avg: 235.6 / Max: 238.42 Min: 224.63 / Avg: 235.03 / Max: 239.68 Min: 231.41 / Avg: 235.98 / Max: 241.38 Min: 234.54 / Avg: 236.34 / Max: 238.73 1. (CC) gcc options: -O2
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 1.59, N = 5 SE +/- 1.54, N = 3 SE +/- 1.26, N = 3 SE +/- 0.95, N = 3 SE +/- 2.03, N = 3 SE +/- 1.01, N = 3 SE +/- 0.89, N = 3 SE +/- 1.58, N = 5 108 109 110 105 109 105 105 108 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 103.92 / Avg: 108.01 / Max: 112.33 Min: 106.92 / Avg: 109.16 / Max: 112.11 Min: 107.75 / Avg: 109.57 / Max: 111.99 Min: 103.4 / Avg: 105.14 / Max: 106.68 Min: 106.27 / Avg: 108.56 / Max: 112.6 Min: 102.87 / Avg: 104.81 / Max: 106.26 Min: 103.14 / Avg: 104.54 / Max: 106.18 Min: 103.83 / Avg: 107.71 / Max: 112.74 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Process GRAPHITE LTO LTO GRAPHITE 0.549 1.098 1.647 2.196 2.745 SE +/- 0.03, N = 3 SE +/- 0.03, N = 6 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 15 SE +/- 0.02, N = 15 SE +/- 0.04, N = 3 SE +/- 0.04, N = 5 2.39 2.34 2.44 2.33 2.41 2.41 2.38 2.35 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Process GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.35 / Avg: 2.39 / Max: 2.45 Min: 2.27 / Avg: 2.34 / Max: 2.5 Min: 2.39 / Avg: 2.44 / Max: 2.53 Min: 2.29 / Avg: 2.33 / Max: 2.36 Min: 2.24 / Avg: 2.41 / Max: 2.59 Min: 2.28 / Avg: 2.41 / Max: 2.61 Min: 2.3 / Avg: 2.38 / Max: 2.44 Min: 2.29 / Avg: 2.35 / Max: 2.5 1. (CC) gcc options: -lpthread -march=native -pipe -O3
System Libxml2 Parsing This test measures the time to parse a random XML file with libxml2 via xmllint using the streaming API. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 112 MB GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 399.30, N = 3 SE +/- 199.91, N = 3 SE +/- 478.48, N = 3 SE +/- 392.00, N = 3 SE +/- 589.51, N = 3 SE +/- 232.90, N = 3 SE +/- 111.92, N = 3 SE +/- 329.72, N = 3 53890 52384 53565 53891 52493 51702 51467 52388 Result Confidence
O2 O3 OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 112 MB GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 53128 / Avg: 53890 / Max: 54478 Min: 52025 / Avg: 52383.67 / Max: 52716 Min: 52921 / Avg: 53565 / Max: 54500 Min: 53264 / Avg: 53890.67 / Max: 54612 Min: 51832 / Avg: 52493 / Max: 53669 Min: 51442 / Avg: 51702.33 / Max: 52167 Min: 51245 / Avg: 51467.33 / Max: 51601 Min: 51761 / Avg: 52388.33 / Max: 52878
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 SE +/- 2.59, N = 3 SE +/- 6.96, N = 3 SE +/- 3.86, N = 9 SE +/- 4.27, N = 8 SE +/- 6.46, N = 3 SE +/- 0.49, N = 3 SE +/- 5.75, N = 3 SE +/- 4.49, N = 7 346 355 343 347 344 359 350 348 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 Min: 341.25 / Avg: 346.41 / Max: 349.38 Min: 341.24 / Avg: 354.63 / Max: 364.63 Min: 333.43 / Avg: 343.02 / Max: 365.48 Min: 325.97 / Avg: 347.16 / Max: 362.16 Min: 332.47 / Avg: 344.17 / Max: 354.75 Min: 358.8 / Avg: 359.38 / Max: 360.35 Min: 341.71 / Avg: 349.6 / Max: 360.79 Min: 325.54 / Avg: 348.17 / Max: 360.9 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 SE +/- 15.17, N = 3 SE +/- 22.95, N = 3 SE +/- 7.00, N = 3 SE +/- 101.13, N = 3 SE +/- 15.81, N = 3 SE +/- 30.79, N = 3 SE +/- 12.50, N = 3 SE +/- 92.48, N = 3 7179 7142 7256 6962 7182 7264 7206 6944 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 Min: 7151.75 / Avg: 7179.35 / Max: 7204.05 Min: 7096.24 / Avg: 7141.74 / Max: 7169.66 Min: 7242.59 / Avg: 7256.11 / Max: 7266.01 Min: 6761.75 / Avg: 6961.92 / Max: 7087.17 Min: 7159.74 / Avg: 7181.88 / Max: 7212.5 Min: 7202.75 / Avg: 7264.32 / Max: 7295.93 Min: 7189.99 / Avg: 7206.14 / Max: 7230.74 Min: 6832.4 / Avg: 6943.63 / Max: 7127.23 1. (CC) gcc options: -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 SE +/- 1.67, N = 3 SE +/- 3.54, N = 3 SE +/- 4.94, N = 3 SE +/- 6.09, N = 3 SE +/- 2.98, N = 3 SE +/- 3.78, N = 3 SE +/- 2.68, N = 3 SE +/- 1.31, N = 3 407 397 392 409 407 410 403 402 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 404.7 / Avg: 407.17 / Max: 410.35 Min: 392.96 / Avg: 396.53 / Max: 403.62 Min: 382.69 / Avg: 392.14 / Max: 399.39 Min: 397.28 / Avg: 408.97 / Max: 417.79 Min: 402.3 / Avg: 407.16 / Max: 412.58 Min: 405.09 / Avg: 409.67 / Max: 417.16 Min: 399.88 / Avg: 402.82 / Max: 408.16 Min: 399.7 / Avg: 402.18 / Max: 404.14 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 5.06, N = 3 SE +/- 0.14, N = 3 SE +/- 1.29, N = 3 SE +/- 2.60, N = 3 SE +/- 3.13, N = 3 SE +/- 4.38, N = 3 SE +/- 13.33, N = 3 SE +/- 6.26, N = 3 1093 1102 1122 1073 1104 1107 1083 1088 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1086.45 / Avg: 1092.66 / Max: 1102.68 Min: 1101.31 / Avg: 1101.55 / Max: 1101.78 Min: 1119.76 / Avg: 1122.16 / Max: 1124.18 Min: 1068.27 / Avg: 1072.95 / Max: 1077.24 Min: 1098.07 / Avg: 1104.13 / Max: 1108.53 Min: 1097.84 / Avg: 1106.57 / Max: 1111.56 Min: 1058.25 / Avg: 1083.28 / Max: 1103.76 Min: 1075.58 / Avg: 1088.03 / Max: 1095.49 1. (CC) gcc options: -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 SE +/- 3.77, N = 15 SE +/- 3.71, N = 15 SE +/- 3.44, N = 15 SE +/- 3.63, N = 15 SE +/- 3.08, N = 15 SE +/- 4.00, N = 15 SE +/- 2.88, N = 15 SE +/- 1.17, N = 3 290 298 289 296 287 293 300 299 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 259.05 / Avg: 290.04 / Max: 313.14 Min: 270.97 / Avg: 298.41 / Max: 316.7 Min: 273.56 / Avg: 289.48 / Max: 320.43 Min: 268.52 / Avg: 296.2 / Max: 318.91 Min: 267.46 / Avg: 287 / Max: 308.18 Min: 271.51 / Avg: 293.23 / Max: 324.7 Min: 273.92 / Avg: 299.72 / Max: 309.71 Min: 297.25 / Avg: 299.26 / Max: 301.31 1. (CC) gcc options: -O2
Himeno Benchmark The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 SE +/- 4.47, N = 3 SE +/- 6.87, N = 3 SE +/- 7.40, N = 3 SE +/- 0.77, N = 3 SE +/- 2.61, N = 3 SE +/- 9.05, N = 3 SE +/- 0.83, N = 3 SE +/- 4.16, N = 3 3284 3292 3412 3408 3276 3297 3400 3424 1. (CC) gcc options: -O3 -march=native -pipe -mavx2 Result Confidence
O2 O3 OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3275.32 / Avg: 3284.24 / Max: 3289.2 Min: 3284.69 / Avg: 3291.68 / Max: 3305.42 Min: 3397.73 / Avg: 3411.72 / Max: 3422.92 Min: 3405.99 / Avg: 3407.54 / Max: 3408.36 Min: 3271.71 / Avg: 3275.63 / Max: 3280.59 Min: 3279.27 / Avg: 3297.34 / Max: 3307.4 Min: 3399.2 / Avg: 3400.46 / Max: 3402.02 Min: 3416.03 / Avg: 3424.31 / Max: 3429.07 1. (CC) gcc options: -O3 -march=native -pipe -mavx2
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 2.08, N = 3 SE +/- 2.91, N = 3 SE +/- 2.98, N = 3 SE +/- 0.98, N = 3 SE +/- 1.33, N = 3 SE +/- 2.85, N = 3 SE +/- 1.07, N = 3 SE +/- 2.01, N = 3 184 181 179 178 180 186 183 184 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 179.57 / Avg: 183.73 / Max: 185.96 Min: 175.4 / Avg: 181.22 / Max: 184.22 Min: 173.69 / Avg: 179.31 / Max: 183.83 Min: 177.03 / Avg: 178.45 / Max: 180.34 Min: 177.35 / Avg: 180.02 / Max: 181.39 Min: 181.87 / Avg: 185.53 / Max: 191.16 Min: 181.62 / Avg: 183.1 / Max: 185.17 Min: 180.11 / Avg: 184.11 / Max: 186.4 1. (CC) gcc options: -O2
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
O2 O3 OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Enhanced GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 258 258 257 258 247 249 248 249 1. (CC) gcc options: -fopenmp -march=native -pipe -O3 -pthread -lwebp -lwebpmux -llcms2 -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lgomp -lpthread
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.45, N = 3 SE +/- 1.53, N = 3 SE +/- 1.45, N = 3 SE +/- 1.20, N = 3 SE +/- 1.45, N = 3 SE +/- 1.67, N = 3 118 114 117 116 115 117 116 113 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 116 / Avg: 118.33 / Max: 121 Min: 112 / Avg: 114 / Max: 117 Min: 115 / Avg: 117.33 / Max: 120 Min: 114 / Avg: 116.33 / Max: 118 Min: 114 / Avg: 116.33 / Max: 119 Min: 111 / Avg: 112.67 / Max: 116 1. (CC) gcc options: -O2 -lpthread
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 15.10, N = 3 SE +/- 17.84, N = 3 SE +/- 5.53, N = 3 SE +/- 5.85, N = 3 SE +/- 14.31, N = 3 SE +/- 6.67, N = 3 SE +/- 6.16, N = 3 SE +/- 4.76, N = 3 1074 1086 1120 1091 1101 1108 1106 1087 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1047.63 / Avg: 1074.2 / Max: 1099.93 Min: 1051.22 / Avg: 1085.83 / Max: 1110.66 Min: 1112.01 / Avg: 1120.02 / Max: 1130.64 Min: 1079.69 / Avg: 1090.67 / Max: 1099.66 Min: 1073.91 / Avg: 1101.1 / Max: 1122.41 Min: 1099.74 / Avg: 1108.22 / Max: 1121.38 Min: 1097.68 / Avg: 1106.09 / Max: 1118.09 Min: 1080.19 / Avg: 1086.66 / Max: 1095.95 1. (CC) gcc options: -O3
Swet Swet is a synthetic CPU/RAM benchmark, includes multi-processor test cases. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Swet 1.5.16 Average GRAPHITE LTO LTO GRAPHITE 200M 400M 600M 800M 1000M SE +/- 7333312.01, N = 3 SE +/- 12426900.31, N = 3 SE +/- 8781499.21, N = 3 SE +/- 3864697.02, N = 3 SE +/- 9725702.11, N = 3 SE +/- 9382367.29, N = 3 SE +/- 6065226.54, N = 3 SE +/- 8667900.17, N = 3 857511567 833713360 822304972 843262796 849927778 842378521 853290616 838587373 1. (CC) gcc options: -lm -lpthread -lcurses -lrt Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Swet 1.5.16 Average GRAPHITE LTO LTO GRAPHITE 150M 300M 450M 600M 750M Min: 843759826 / Avg: 857511566.67 / Max: 868803274 Min: 817689063 / Avg: 833713359.67 / Max: 858178471 Min: 806905975 / Avg: 822304972 / Max: 837318659 Min: 835537836 / Avg: 843262796.33 / Max: 847351970 Min: 831285654 / Avg: 849927777.67 / Max: 864057297 Min: 825846066 / Avg: 842378521 / Max: 858332069 Min: 845050797 / Avg: 853290616.33 / Max: 865120257 Min: 826224550 / Avg: 838587372.67 / Max: 855293419 1. (CC) gcc options: -lm -lpthread -lcurses -lrt
Tachyon This is a test of the threaded Tachyon, a parallel ray-tracing system. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.98.9 Total Time GRAPHITE LTO LTO GRAPHITE 0.4388 0.8776 1.3164 1.7552 2.194 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 7 SE +/- 0.02, N = 3 SE +/- 0.03, N = 4 SE +/- 0.04, N = 3 1.92 1.87 1.91 1.89 1.95 1.91 1.91 1.89 1. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.98.9 Total Time GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.88 / Avg: 1.92 / Max: 1.94 Min: 1.84 / Avg: 1.87 / Max: 1.88 Min: 1.86 / Avg: 1.91 / Max: 1.95 Min: 1.86 / Avg: 1.89 / Max: 1.92 Min: 1.84 / Avg: 1.95 / Max: 2.04 Min: 1.89 / Avg: 1.91 / Max: 1.95 Min: 1.84 / Avg: 1.91 / Max: 1.98 Min: 1.83 / Avg: 1.89 / Max: 1.95 1. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 25.66, N = 4 SE +/- 41.76, N = 4 SE +/- 18.82, N = 4 SE +/- 22.76, N = 16 SE +/- 15.26, N = 4 SE +/- 17.03, N = 4 SE +/- 32.95, N = 7 SE +/- 39.00, N = 4 2628 2526 2526 2622 2634 2546 2606 2609 Result Confidence
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 Min: 2567 / Avg: 2627.75 / Max: 2688 Min: 2473 / Avg: 2526.25 / Max: 2650 Min: 2475 / Avg: 2525.5 / Max: 2564 Min: 2471 / Avg: 2621.81 / Max: 2793 Min: 2600 / Avg: 2633.75 / Max: 2666 Min: 2497 / Avg: 2546.25 / Max: 2573 Min: 2476 / Avg: 2606 / Max: 2729 Min: 2547 / Avg: 2608.75 / Max: 2713
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 900 1800 2700 3600 4500 SE +/- 63.91, N = 3 SE +/- 48.64, N = 3 SE +/- 17.59, N = 3 SE +/- 28.49, N = 3 SE +/- 14.45, N = 3 SE +/- 61.71, N = 3 SE +/- 61.11, N = 3 SE +/- 68.23, N = 3 4171 4161 4246 4148 4097 4272 4263 4139 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 4092.36 / Avg: 4170.83 / Max: 4297.44 Min: 4064.17 / Avg: 4161.45 / Max: 4210.98 Min: 4219.69 / Avg: 4245.88 / Max: 4279.32 Min: 4093.51 / Avg: 4147.79 / Max: 4189.95 Min: 4068.14 / Avg: 4097.01 / Max: 4112.58 Min: 4148.57 / Avg: 4271.75 / Max: 4340.11 Min: 4140.82 / Avg: 4263.05 / Max: 4324.8 Min: 4061.43 / Avg: 4139.08 / Max: 4275.08 1. (CC) gcc options: -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 492.06, N = 3 SE +/- 601.89, N = 3 SE +/- 338.66, N = 3 SE +/- 153.84, N = 3 SE +/- 256.00, N = 3 SE +/- 619.77, N = 3 SE +/- 298.67, N = 3 SE +/- 85.33, N = 3 49003 48491 48960 49771 48960 50539 48917 48789 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 48448 / Avg: 49002.67 / Max: 49984 Min: 47296 / Avg: 48490.67 / Max: 49216 Min: 48448 / Avg: 48960 / Max: 49600 Min: 49472 / Avg: 49770.67 / Max: 49984 Min: 48704 / Avg: 48960 / Max: 49472 Min: 49856 / Avg: 50538.67 / Max: 51776 Min: 48320 / Avg: 48917.33 / Max: 49216 Min: 48704 / Avg: 48789.33 / Max: 48960 1. (CC) gcc options: -O2 -lpthread
CLOMP CLOMP is the C version of the Livermore OpenMP benchmark developed to measure OpenMP overheads and other performance impacts due to threading in order to influence future system designs. This particular test profile configuration is currently set to look at the OpenMP static schedule speed-up across all available CPU cores using the recommended test configuration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Speedup, More Is Better CLOMP 3.3 Static OMP Speedup GRAPHITE LTO LTO GRAPHITE 0.3915 0.783 1.1745 1.566 1.9575 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 1.71 1.67 1.72 1.70 1.74 1.69 1.69 1.70 1. (CC) gcc options: --openmp -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Speedup, More Is Better CLOMP 3.3 Static OMP Speedup GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.68 / Avg: 1.71 / Max: 1.75 Min: 1.64 / Avg: 1.67 / Max: 1.71 Min: 1.67 / Avg: 1.72 / Max: 1.79 Min: 1.68 / Avg: 1.7 / Max: 1.72 Min: 1.69 / Avg: 1.74 / Max: 1.81 Min: 1.67 / Avg: 1.69 / Max: 1.73 Min: 1.65 / Avg: 1.69 / Max: 1.75 Min: 1.67 / Avg: 1.7 / Max: 1.73 1. (CC) gcc options: --openmp -O3 -lm
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh GRAPHITE LTO LTO GRAPHITE 0.225 0.45 0.675 0.9 1.125 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.00 0.99 0.97 0.96 0.99 1.00 0.97 0.97 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1 / Avg: 1 / Max: 1 Min: 0.99 / Avg: 0.99 / Max: 0.99 Min: 0.97 / Avg: 0.97 / Max: 0.97 Min: 0.96 / Avg: 0.96 / Max: 0.97 Min: 0.99 / Avg: 0.99 / Max: 1 Min: 1 / Avg: 1 / Max: 1 Min: 0.97 / Avg: 0.97 / Max: 0.97 Min: 0.96 / Avg: 0.97 / Max: 0.97 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 SE +/- 2.77, N = 3 SE +/- 3.40, N = 3 SE +/- 3.98, N = 3 SE +/- 0.29, N = 3 SE +/- 4.88, N = 3 SE +/- 5.22, N = 3 SE +/- 2.64, N = 3 SE +/- 5.96, N = 3 324 322 321 320 325 327 325 314 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 Min: 319.46 / Avg: 323.99 / Max: 329.03 Min: 317.14 / Avg: 321.64 / Max: 328.3 Min: 315.63 / Avg: 320.52 / Max: 328.42 Min: 319 / Avg: 319.58 / Max: 319.91 Min: 316.11 / Avg: 325.12 / Max: 332.88 Min: 321.07 / Avg: 326.64 / Max: 337.07 Min: 321.39 / Avg: 324.82 / Max: 330.01 Min: 304.16 / Avg: 314.15 / Max: 324.78 1. (CC) gcc options: -O2
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 10.75, N = 3 SE +/- 14.87, N = 3 SE +/- 4.39, N = 3 SE +/- 2.09, N = 3 SE +/- 0.95, N = 3 SE +/- 0.79, N = 3 SE +/- 2.64, N = 3 SE +/- 0.41, N = 3 1207 1223 1253 1229 1250 1256 1255 1231 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1185.97 / Avg: 1207.43 / Max: 1219.26 Min: 1196.76 / Avg: 1222.67 / Max: 1248.26 Min: 1246.25 / Avg: 1252.53 / Max: 1260.99 Min: 1225.61 / Avg: 1229.05 / Max: 1232.82 Min: 1248.25 / Avg: 1250.13 / Max: 1251.33 Min: 1254.12 / Avg: 1255.67 / Max: 1256.7 Min: 1249.62 / Avg: 1254.9 / Max: 1257.79 Min: 1230.68 / Avg: 1231.44 / Max: 1232.08 1. (CC) gcc options: -O3
Sockperf This is a network socket API performance benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput GRAPHITE LTO LTO GRAPHITE 120K 240K 360K 480K 600K SE +/- 6878.12, N = 5 SE +/- 5296.28, N = 5 SE +/- 2569.74, N = 5 SE +/- 5657.41, N = 5 SE +/- 1934.24, N = 5 SE +/- 7472.89, N = 5 SE +/- 3175.67, N = 5 SE +/- 5461.64, N = 5 529960 539098 531031 532851 535609 518224 527941 522348 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread Result Confidence
O2 O3 OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput GRAPHITE LTO LTO GRAPHITE 90K 180K 270K 360K 450K Min: 503701 / Avg: 529959.6 / Max: 541680 Min: 521338 / Avg: 539097.8 / Max: 552565 Min: 521932 / Avg: 531030.8 / Max: 537805 Min: 514980 / Avg: 532851.2 / Max: 543083 Min: 529006 / Avg: 535609 / Max: 540214 Min: 500191 / Avg: 518224 / Max: 538031 Min: 517518 / Avg: 527940.6 / Max: 535698 Min: 513317 / Avg: 522347.6 / Max: 543292 1. (CXX) g++ options: --param -march=native -pipe -O3 -rdynamic -ldl -lpthread
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 1.20, N = 3 SE +/- 1.20, N = 3 SE +/- 0.58, N = 3 102 105 104 106 105 104 104 105 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 101 / Avg: 102.33 / Max: 104 Min: 104 / Avg: 105 / Max: 106 Min: 103 / Avg: 104 / Max: 105 Min: 102 / Avg: 103.67 / Max: 106 Min: 102 / Avg: 103.67 / Max: 106 Min: 104 / Avg: 105 / Max: 106 1. (CC) gcc options: -O2 -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 105.09, N = 3 SE +/- 104.48, N = 3 SE +/- 130.08, N = 3 SE +/- 169.24, N = 3 SE +/- 84.08, N = 3 SE +/- 113.98, N = 3 SE +/- 44.99, N = 3 SE +/- 151.80, N = 3 14940 14881 14632 14870 14951 15027 15203 15083 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 14744.85 / Avg: 14940.21 / Max: 15105.03 Min: 14754.19 / Avg: 14880.85 / Max: 15088.11 Min: 14468.18 / Avg: 14631.98 / Max: 14888.92 Min: 14576.93 / Avg: 14870.29 / Max: 15163.18 Min: 14799.4 / Avg: 14950.73 / Max: 15089.89 Min: 14801.97 / Avg: 15026.68 / Max: 15172.17 Min: 15113.05 / Avg: 15202.77 / Max: 15253.53 Min: 14843.16 / Avg: 15083.45 / Max: 15364.3 1. (CC) gcc options: -O2
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
O2 O3 OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Sharpen GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 240 239 239 240 231 232 232 232 1. (CC) gcc options: -fopenmp -march=native -pipe -O3 -pthread -lwebp -lwebpmux -llcms2 -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lgomp -lpthread
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Composite GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 SE +/- 27.36, N = 4 SE +/- 26.17, N = 4 SE +/- 23.90, N = 4 SE +/- 4.59, N = 4 SE +/- 31.58, N = 4 SE +/- 26.00, N = 4 SE +/- 27.43, N = 4 SE +/- 26.78, N = 4 2297 2310 2308 2224 2268 2273 2250 2242 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Composite GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 2215.47 / Avg: 2297.37 / Max: 2329.8 Min: 2231.23 / Avg: 2309.71 / Max: 2338.25 Min: 2236.68 / Avg: 2307.62 / Max: 2340.77 Min: 2215.02 / Avg: 2224.36 / Max: 2233.55 Min: 2209.4 / Avg: 2267.8 / Max: 2326.13 Min: 2218.73 / Avg: 2273.32 / Max: 2320.49 Min: 2208.76 / Avg: 2249.57 / Max: 2330.38 Min: 2206.87 / Avg: 2241.8 / Max: 2321.01
Sysbench This is a benchmark of Sysbench with CPU and memory sub-tests. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: CPU GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 4.97, N = 3 SE +/- 5.60, N = 3 SE +/- 1.97, N = 3 SE +/- 1.46, N = 3 SE +/- 2.72, N = 3 SE +/- 7.30, N = 3 SE +/- 5.61, N = 3 SE +/- 7.45, N = 3 42247 42247 43857 43867 43866 42283 43693 42259 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=native -pipe -rdynamic -ldl -laio -lm Result Confidence
O2 O3 OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: CPU GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K Min: 42237.92 / Avg: 42247.24 / Max: 42254.87 Min: 42235.68 / Avg: 42246.88 / Max: 42252.63 Min: 43854.44 / Avg: 43857.11 / Max: 43860.96 Min: 43863.7 / Avg: 43866.58 / Max: 43868.45 Min: 43860.57 / Avg: 43865.56 / Max: 43869.95 Min: 42275.92 / Avg: 42283.37 / Max: 42297.97 Min: 43686.35 / Avg: 43693.08 / Max: 43704.23 Min: 42244.56 / Avg: 42258.74 / Max: 42269.77 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=native -pipe -rdynamic -ldl -laio -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 2.00, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 1.20, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 107 106 106 108 105 106 109 105 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 105 / Avg: 107 / Max: 111 Min: 105 / Avg: 106 / Max: 107 Min: 105 / Avg: 106.33 / Max: 108 Min: 106 / Avg: 107.67 / Max: 110 Min: 108 / Avg: 109 / Max: 110 Min: 104 / Avg: 105.33 / Max: 107 1. (CC) gcc options: -O2 -lpthread
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Delete GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 785.71, N = 3 SE +/- 357.66, N = 3 SE +/- 314.11, N = 3 SE +/- 273.03, N = 3 SE +/- 266.31, N = 3 SE +/- 1059.73, N = 3 SE +/- 479.41, N = 3 SE +/- 214.88, N = 3 112024 112235 111069 112297 113836 110707 114217 114912 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Delete GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 110842.2 / Avg: 112024.47 / Max: 113512.1 Min: 111530.9 / Avg: 112234.57 / Max: 112697.8 Min: 110683.4 / Avg: 111068.5 / Max: 111690.9 Min: 111933.8 / Avg: 112297.37 / Max: 112832 Min: 113348.8 / Avg: 113835.77 / Max: 114266.1 Min: 109580.2 / Avg: 110706.77 / Max: 112824.8 Min: 113472 / Avg: 114217.2 / Max: 115112.3 Min: 114560.3 / Avg: 114912.23 / Max: 115301.8 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 60K 120K 180K 240K 300K SE +/- 3876.80, N = 3 SE +/- 4827.94, N = 4 SE +/- 2665.90, N = 3 SE +/- 2995.38, N = 10 SE +/- 1806.17, N = 3 SE +/- 3563.62, N = 3 SE +/- 2238.27, N = 3 SE +/- 1564.19, N = 3 273920 278272 279723 276275 281771 271531 274603 271872 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 50K 100K 150K 200K 250K Min: 266752 / Avg: 273920 / Max: 280064 Min: 268800 / Avg: 278272 / Max: 291328 Min: 274944 / Avg: 279722.67 / Max: 284160 Min: 266752 / Avg: 276275.2 / Max: 299520 Min: 279040 / Avg: 281770.67 / Max: 285184 Min: 265728 / Avg: 271530.67 / Max: 278016 Min: 271872 / Avg: 274602.67 / Max: 279040 Min: 269824 / Avg: 271872 / Max: 274944 1. (CC) gcc options: -O2 -lpthread
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 110 220 330 440 550 SE +/- 1.01, N = 3 SE +/- 7.57, N = 3 SE +/- 11.73, N = 3 SE +/- 4.52, N = 3 SE +/- 0.30, N = 3 SE +/- 4.65, N = 3 SE +/- 8.27, N = 3 SE +/- 6.26, N = 3 518 511 512 504 511 523 516 506 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 Min: 516.32 / Avg: 517.83 / Max: 519.74 Min: 495.8 / Avg: 510.94 / Max: 518.96 Min: 488.31 / Avg: 511.71 / Max: 524.83 Min: 495.3 / Avg: 504.25 / Max: 509.85 Min: 510 / Avg: 510.58 / Max: 510.97 Min: 513.93 / Avg: 523.22 / Max: 528.19 Min: 499.45 / Avg: 515.95 / Max: 525.3 Min: 493.71 / Avg: 506.01 / Max: 514.2 1. (CC) gcc options: -lm -ldl
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K SE +/- 27135.03, N = 4 SE +/- 6294.27, N = 3 SE +/- 15081.27, N = 3 SE +/- 11761.89, N = 3 SE +/- 27774.86, N = 3 SE +/- 14392.39, N = 3 SE +/- 7641.91, N = 3 SE +/- 20329.70, N = 15 1615437 1636485 1628043 1626213 1603673 1618376 1620539 1577352 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K Min: 1535808.02 / Avg: 1615437.38 / Max: 1650282.03 Min: 1624752.08 / Avg: 1636484.78 / Max: 1646302.25 Min: 1608166.47 / Avg: 1628042.55 / Max: 1657628.49 Min: 1603131.12 / Avg: 1626213.26 / Max: 1641683.37 Min: 1556822.5 / Avg: 1603673.13 / Max: 1652945.92 Min: 1593154.42 / Avg: 1618375.72 / Max: 1643000.4 Min: 1611963.79 / Avg: 1620538.58 / Max: 1635782.75 Min: 1368983.96 / Avg: 1577351.62 / Max: 1658299.6 1. (CC) gcc options: -O2
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: BT.A GRAPHITE LTO LTO GRAPHITE 1200 2400 3600 4800 6000 SE +/- 20.73, N = 3 SE +/- 28.65, N = 3 SE +/- 17.41, N = 3 SE +/- 20.16, N = 3 SE +/- 41.72, N = 3 SE +/- 10.93, N = 3 SE +/- 12.26, N = 3 SE +/- 7.36, N = 3 5544 5449 5631 5492 5530 5430 5626 5548 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: BT.A GRAPHITE LTO LTO GRAPHITE 1000 2000 3000 4000 5000 Min: 5503.46 / Avg: 5544.38 / Max: 5570.62 Min: 5411.21 / Avg: 5449.44 / Max: 5505.52 Min: 5595.77 / Avg: 5630.51 / Max: 5649.86 Min: 5465.5 / Avg: 5491.99 / Max: 5531.56 Min: 5457.75 / Avg: 5529.88 / Max: 5602.27 Min: 5408.91 / Avg: 5429.76 / Max: 5445.89 Min: 5606.42 / Avg: 5625.94 / Max: 5648.54 Min: 5538.97 / Avg: 5547.83 / Max: 5562.44 1. 3.3
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 4.77, N = 3 SE +/- 6.15, N = 3 SE +/- 7.74, N = 3 SE +/- 6.64, N = 3 SE +/- 3.57, N = 3 SE +/- 3.53, N = 3 SE +/- 6.14, N = 3 SE +/- 10.84, N = 3 1207 1215 1201 1243 1228 1238 1199 1240 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1198.97 / Avg: 1206.84 / Max: 1215.45 Min: 1208.71 / Avg: 1215.28 / Max: 1227.56 Min: 1185.87 / Avg: 1201.35 / Max: 1209.4 Min: 1229.53 / Avg: 1242.54 / Max: 1251.32 Min: 1221.94 / Avg: 1228.03 / Max: 1234.3 Min: 1233.12 / Avg: 1237.8 / Max: 1244.73 Min: 1189.12 / Avg: 1199.15 / Max: 1210.31 Min: 1218.59 / Avg: 1239.91 / Max: 1254.02 1. (CC) gcc options: -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K SE +/- 2172.30, N = 15 SE +/- 2917.80, N = 15 SE +/- 2028.18, N = 15 SE +/- 2795.74, N = 5 SE +/- 3723.49, N = 15 SE +/- 2097.67, N = 3 SE +/- 3648.75, N = 15 SE +/- 2819.60, N = 15 190134 188525 192510 192841 189083 195217 192702 193239 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K Min: 171927.47 / Avg: 190133.99 / Max: 200312.99 Min: 159900.06 / Avg: 188524.88 / Max: 204146.73 Min: 168089.3 / Avg: 192509.81 / Max: 202692 Min: 187545.79 / Avg: 192840.54 / Max: 203659.51 Min: 161209.07 / Avg: 189082.88 / Max: 210353.33 Min: 191044.78 / Avg: 195217.08 / Max: 197683.4 Min: 162127.93 / Avg: 192701.76 / Max: 206119.16 Min: 161514.2 / Avg: 193238.92 / Max: 210180.62 1. (CC) gcc options: -O2
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Random Numbers GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 2.43, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 990 989 959 993 988 989 959 993 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Random Numbers GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 987.99 / Avg: 990.49 / Max: 995.35 Min: 988.54 / Avg: 988.59 / Max: 988.62 Min: 958.75 / Avg: 958.8 / Max: 958.9 Min: 992.78 / Avg: 992.84 / Max: 992.9 Min: 988.21 / Avg: 988.35 / Max: 988.62 Min: 988.5 / Avg: 988.54 / Max: 988.63 Min: 958.7 / Avg: 958.71 / Max: 958.71 Min: 992.82 / Avg: 992.96 / Max: 993.08 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 610.80, N = 3 SE +/- 449.34, N = 6 SE +/- 407.01, N = 3 SE +/- 448.16, N = 5 SE +/- 570.14, N = 3 SE +/- 439.67, N = 3 SE +/- 594.66, N = 3 SE +/- 605.75, N = 3 31808 32117 32053 32634 32523 32405 31520 32128 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 30880 / Avg: 31808 / Max: 32960 Min: 30432 / Avg: 32117.33 / Max: 32960 Min: 31328 / Avg: 32053.33 / Max: 32736 Min: 31328 / Avg: 32633.6 / Max: 34112 Min: 31392 / Avg: 32522.67 / Max: 33216 Min: 31584 / Avg: 32405.33 / Max: 33088 Min: 30560 / Avg: 31520 / Max: 32608 Min: 31008 / Avg: 32128 / Max: 33088 1. (CC) gcc options: -O2 -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 200K 400K 600K 800K 1000K SE +/- 10849.39, N = 3 SE +/- 9767.94, N = 15 SE +/- 8610.55, N = 3 SE +/- 11379.65, N = 15 SE +/- 8941.82, N = 9 SE +/- 11576.21, N = 3 SE +/- 2639.63, N = 3 SE +/- 8131.18, N = 15 821454 808452 817530 793555 802253 814108 798043 798177 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 140K 280K 420K 560K 700K Min: 810126.58 / Avg: 821453.67 / Max: 843145.33 Min: 689330.19 / Avg: 808451.84 / Max: 860142.8 Min: 800938.6 / Avg: 817530.06 / Max: 829821.72 Min: 662997.73 / Avg: 793555.41 / Max: 823151.13 Min: 747991.23 / Avg: 802252.83 / Max: 828143.96 Min: 791344.67 / Avg: 814108.11 / Max: 829149.8 Min: 792876.5 / Avg: 798043.23 / Max: 801565.56 Min: 732474.96 / Avg: 798177.42 / Max: 847682.12 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 700K 1400K 2100K 2800K 3500K SE +/- 44001.81, N = 3 SE +/- 32190.17, N = 3 SE +/- 39307.38, N = 12 SE +/- 24286.71, N = 3 SE +/- 51183.08, N = 15 SE +/- 38518.43, N = 8 SE +/- 10648.00, N = 3 SE +/- 55839.58, N = 3 3180525 3238580 3207864 3213337 3138260 3186499 3163820 3129139 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 600K 1200K 1800K 2400K 3000K Min: 3099508.14 / Avg: 3180525.22 / Max: 3250793.65 Min: 3192517.54 / Avg: 3238580.31 / Max: 3300564.06 Min: 2820936.64 / Avg: 3207864.14 / Max: 3357377.05 Min: 3167826.76 / Avg: 3213337.49 / Max: 3250793.65 Min: 2674502.12 / Avg: 3138260.49 / Max: 3315256.98 Min: 2936200.72 / Avg: 3186498.85 / Max: 3311236.86 Min: 3151981.53 / Avg: 3163820.15 / Max: 3185069.98 Min: 3019535.57 / Avg: 3129138.73 / Max: 3202501.96 1. (CC) gcc options: -O2
Git This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands GRAPHITE LTO LTO GRAPHITE 1.0778 2.1556 3.2334 4.3112 5.389 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 4.69 4.79 4.67 4.73 4.72 4.75 4.63 4.70 1. git version 2.22.1 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 4.64 / Avg: 4.69 / Max: 4.75 Min: 4.79 / Avg: 4.79 / Max: 4.79 Min: 4.61 / Avg: 4.67 / Max: 4.77 Min: 4.66 / Avg: 4.73 / Max: 4.8 Min: 4.6 / Avg: 4.72 / Max: 4.78 Min: 4.7 / Avg: 4.75 / Max: 4.79 Min: 4.5 / Avg: 4.63 / Max: 4.69 Min: 4.7 / Avg: 4.7 / Max: 4.71 1. git version 2.22.1
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire GRAPHITE LTO LTO GRAPHITE 15 30 45 60 75 SE +/- 0.52, N = 3 SE +/- 1.04, N = 3 SE +/- 0.74, N = 3 SE +/- 0.88, N = 3 SE +/- 0.92, N = 3 SE +/- 1.27, N = 3 SE +/- 1.18, N = 3 SE +/- 0.44, N = 3 65.93 65.33 64.70 66.76 65.15 65.48 65.62 64.55 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 Min: 65.03 / Avg: 65.93 / Max: 66.82 Min: 64.22 / Avg: 65.33 / Max: 67.4 Min: 63.88 / Avg: 64.7 / Max: 66.18 Min: 65.25 / Avg: 66.76 / Max: 68.29 Min: 63.63 / Avg: 65.15 / Max: 66.81 Min: 63.58 / Avg: 65.48 / Max: 67.88 Min: 64.26 / Avg: 65.62 / Max: 67.98 Min: 63.67 / Avg: 64.55 / Max: 65.03 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.33, N = 3 SE +/- 2.08, N = 3 SE +/- 1.45, N = 3 SE +/- 1.86, N = 3 SE +/- 1.67, N = 3 SE +/- 2.08, N = 3 SE +/- 1.67, N = 6 SE +/- 0.58, N = 3 119 120 117 119 119 121 118 117 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 118 / Avg: 119.33 / Max: 122 Min: 116 / Avg: 120 / Max: 123 Min: 114 / Avg: 116.67 / Max: 119 Min: 117 / Avg: 119.33 / Max: 123 Min: 116 / Avg: 119.33 / Max: 121 Min: 118 / Avg: 121 / Max: 125 Min: 115 / Avg: 118.33 / Max: 126 Min: 116 / Avg: 117 / Max: 118 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K SE +/- 73.90, N = 3 SE +/- 349.24, N = 3 SE +/- 225.77, N = 3 SE +/- 592.74, N = 3 SE +/- 298.67, N = 3 SE +/- 449.52, N = 3 SE +/- 407.01, N = 3 44352 44608 44011 44779 44437 44907 45504 44395 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K Min: 44224 / Avg: 44352 / Max: 44480 Min: 43328 / Avg: 44010.67 / Max: 44480 Min: 44352 / Avg: 44778.67 / Max: 45120 Min: 43456 / Avg: 44437.33 / Max: 45504 Min: 44352 / Avg: 44906.67 / Max: 45376 Min: 44608 / Avg: 45504 / Max: 46016 Min: 43584 / Avg: 44394.67 / Max: 44864 1. (CC) gcc options: -O2 -lpthread
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.53, N = 3 SE +/- 2.03, N = 3 SE +/- 2.00, N = 3 SE +/- 2.03, N = 3 SE +/- 1.53, N = 3 SE +/- 1.97, N = 4 SE +/- 1.81, N = 5 SE +/- 2.08, N = 3 122 119 119 122 123 122 121 119 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 120 / Avg: 122 / Max: 125 Min: 115 / Avg: 118.67 / Max: 122 Min: 117 / Avg: 119 / Max: 123 Min: 119 / Avg: 122.33 / Max: 126 Min: 121 / Avg: 123 / Max: 126 Min: 116 / Avg: 121.75 / Max: 125 Min: 117 / Avg: 121.4 / Max: 128 Min: 115 / Avg: 119 / Max: 122 1. (CC) gcc options: -O2 -lpthread
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 SE +/- 29.20, N = 3 SE +/- 22.91, N = 3 SE +/- 49.79, N = 3 SE +/- 11.76, N = 3 SE +/- 30.82, N = 3 SE +/- 21.38, N = 3 SE +/- 34.52, N = 3 SE +/- 49.45, N = 3 7615 7474 7672 7425 7512 7648 7518 7497 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 Min: 7580.45 / Avg: 7614.76 / Max: 7672.85 Min: 7435.33 / Avg: 7473.7 / Max: 7514.56 Min: 7618.23 / Avg: 7672.49 / Max: 7771.93 Min: 7407.35 / Avg: 7425.48 / Max: 7447.53 Min: 7463.81 / Avg: 7511.75 / Max: 7569.26 Min: 7622.04 / Avg: 7647.97 / Max: 7690.38 Min: 7456.53 / Avg: 7518.39 / Max: 7575.86 Min: 7399.69 / Avg: 7497.21 / Max: 7560.21 1. (CC) gcc options: -O3
FS-Mark FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 9.40, N = 2 SE +/- 5.79, N = 3 SE +/- 12.20, N = 3 SE +/- 5.00, N = 3 SE +/- 3.01, N = 3 SE +/- 4.89, N = 3 SE +/- 5.66, N = 3 SE +/- 1.39, N = 3 1519 1513 1515 1546 1563 1533 1526 1527 1. (CC) gcc options: -static Result Confidence
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 Min: 1509.2 / Avg: 1518.6 / Max: 1528 Min: 1502.8 / Avg: 1513.2 / Max: 1522.8 Min: 1491.2 / Avg: 1514.8 / Max: 1532 Min: 1541.2 / Avg: 1546.4 / Max: 1556.4 Min: 1557.6 / Avg: 1562.67 / Max: 1568 Min: 1523.2 / Avg: 1532.8 / Max: 1539.2 Min: 1516.8 / Avg: 1526.4 / Max: 1536.4 Min: 1524.8 / Avg: 1527.2 / Max: 1529.6 1. (CC) gcc options: -static
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 1230.69, N = 3 SE +/- 256.00, N = 3 SE +/- 1257.04, N = 3 SE +/- 1257.04, N = 3 SE +/- 698.48, N = 3 SE +/- 391.05, N = 3 SE +/- 532.91, N = 3 SE +/- 451.54, N = 3 75563 77952 75477 76587 76331 77440 76160 75477 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 14K 28K 42K 56K 70K Min: 73856 / Avg: 75562.67 / Max: 77952 Min: 77696 / Avg: 77952 / Max: 78464 Min: 73856 / Avg: 75477.33 / Max: 77952 Min: 74368 / Avg: 76586.67 / Max: 78720 Min: 75392 / Avg: 76330.67 / Max: 77696 Min: 76672 / Avg: 77440 / Max: 77952 Min: 75136 / Avg: 76160 / Max: 76928 Min: 74624 / Avg: 75477.33 / Max: 76160 1. (CC) gcc options: -O2 -lpthread
BlogBench BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read GRAPHITE LTO LTO GRAPHITE 800K 1600K 2400K 3200K 4000K SE +/- 40313.06, N = 3 SE +/- 33892.32, N = 3 SE +/- 35356.74, N = 3 SE +/- 21282.43, N = 3 SE +/- 75432.58, N = 3 SE +/- 24300.12, N = 3 SE +/- 19918.02, N = 3 SE +/- 5417.42, N = 3 3763979 3840238 3831194 3819147 3878929 3781351 3865835 3886262 1. (CC) gcc options: -march=native -pipe -O3 -pthread Result Confidence
O2 O3 OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read GRAPHITE LTO LTO GRAPHITE 700K 1400K 2100K 2800K 3500K Min: 3690271 / Avg: 3763979 / Max: 3829131 Min: 3794437 / Avg: 3840238 / Max: 3906414 Min: 3765740 / Avg: 3831194 / Max: 3887097 Min: 3796989 / Avg: 3819147.33 / Max: 3861700 Min: 3771133 / Avg: 3878929 / Max: 4024234 Min: 3737931 / Avg: 3781351.33 / Max: 3821969 Min: 3827890 / Avg: 3865834.67 / Max: 3895311 Min: 3876439 / Avg: 3886261.67 / Max: 3895133 1. (CC) gcc options: -march=native -pipe -O3 -pthread
AIO-Stress AIO-Stress is an a-synchronous I/O benchmark created by SuSE. Current this profile uses a 2048MB test file and a 64KB record size. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better AIO-Stress 0.21 Test: Random Write GRAPHITE LTO LTO GRAPHITE 900 1800 2700 3600 4500 SE +/- 7.74, N = 3 SE +/- 4.44, N = 3 SE +/- 5.89, N = 3 SE +/- 7.99, N = 3 SE +/- 7.20, N = 3 SE +/- 2.68, N = 3 SE +/- 4.81, N = 3 SE +/- 7.35, N = 3 3973 4040 4101 3979 4051 3988 3984 3983 1. (CC) gcc options: -pthread -laio Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better AIO-Stress 0.21 Test: Random Write GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 3964.04 / Avg: 3972.68 / Max: 3988.12 Min: 4031.27 / Avg: 4039.89 / Max: 4046.02 Min: 4090.6 / Avg: 4101.44 / Max: 4110.84 Min: 3970.44 / Avg: 3979.32 / Max: 3995.27 Min: 4043.52 / Avg: 4051.19 / Max: 4065.59 Min: 3983.41 / Avg: 3987.57 / Max: 3992.59 Min: 3976.85 / Avg: 3984.18 / Max: 3993.24 Min: 3971.99 / Avg: 3982.69 / Max: 3996.76 1. (CC) gcc options: -pthread -laio
O2 O3 OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Floating Point GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K 32076 32163 31951 31521 31458 32423 31414 31700 1. (CC) gcc options: -O3 -march=native -pipe
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 615.35, N = 3 SE +/- 364.54, N = 3 SE +/- 147.80, N = 3 SE +/- 42.67, N = 3 SE +/- 371.96, N = 3 SE +/- 762.22, N = 5 SE +/- 666.47, N = 3 54507 55787 55744 56256 56043 55275 54925 54891 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 53312 / Avg: 54506.67 / Max: 55360 Min: 55360 / Avg: 55786.67 / Max: 56512 Min: 56000 / Avg: 56256 / Max: 56512 Min: 56000 / Avg: 56042.67 / Max: 56128 Min: 54592 / Avg: 55274.67 / Max: 55872 Min: 52544 / Avg: 54924.8 / Max: 57280 Min: 53696 / Avg: 54890.67 / Max: 56000 1. (CC) gcc options: -O2 -lpthread
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 400.40, N = 3 SE +/- 226.59, N = 3 SE +/- 280.15, N = 3 SE +/- 43.68, N = 3 SE +/- 33.41, N = 3 SE +/- 155.11, N = 3 SE +/- 74.75, N = 3 SE +/- 338.06, N = 3 24162 24696 24332 24490 24654 23967 24468 23933 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 23385 / Avg: 24161.67 / Max: 24719 Min: 24247 / Avg: 24696.33 / Max: 24972 Min: 23773 / Avg: 24331.67 / Max: 24648 Min: 24405 / Avg: 24490 / Max: 24550 Min: 24587 / Avg: 24653.67 / Max: 24691 Min: 23675 / Avg: 23966.67 / Max: 24204 Min: 24364 / Avg: 24468 / Max: 24613 Min: 23321 / Avg: 23932.67 / Max: 24488 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 4.35, N = 3 SE +/- 2.14, N = 3 SE +/- 4.35, N = 3 SE +/- 2.87, N = 3 SE +/- 9.43, N = 3 SE +/- 4.00, N = 3 SE +/- 5.86, N = 3 SE +/- 12.52, N = 3 1083 1088 1068 1100 1083 1096 1074 1102 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1078.1 / Avg: 1082.95 / Max: 1091.63 Min: 1083.76 / Avg: 1087.78 / Max: 1091.06 Min: 1063.14 / Avg: 1067.53 / Max: 1076.22 Min: 1094.24 / Avg: 1099.96 / Max: 1103.31 Min: 1073.34 / Avg: 1083.18 / Max: 1102.04 Min: 1089.46 / Avg: 1095.55 / Max: 1103.08 Min: 1064.19 / Avg: 1073.65 / Max: 1084.38 Min: 1077.47 / Avg: 1102.46 / Max: 1116.28 1. (CC) gcc options: -O3
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 5.23, N = 3 SE +/- 6.54, N = 3 SE +/- 1.24, N = 3 SE +/- 2.49, N = 3 SE +/- 0.53, N = 3 SE +/- 6.58, N = 3 SE +/- 1.90, N = 3 SE +/- 0.67, N = 3 1222 1232 1233 1213 1205 1243 1238 1226 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1211.98 / Avg: 1222.09 / Max: 1229.49 Min: 1219.15 / Avg: 1232.22 / Max: 1238.82 Min: 1230.61 / Avg: 1233.07 / Max: 1234.55 Min: 1208.86 / Avg: 1213.11 / Max: 1217.5 Min: 1204.21 / Avg: 1205.17 / Max: 1206.04 Min: 1229.73 / Avg: 1242.62 / Max: 1251.35 Min: 1235.7 / Avg: 1238.15 / Max: 1241.88 Min: 1224.65 / Avg: 1225.99 / Max: 1226.79 1. (CC) gcc options: -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 426.61, N = 4 SE +/- 457.54, N = 3 SE +/- 325.62, N = 3 SE +/- 37.99, N = 3 SE +/- 413.09, N = 3 SE +/- 122.88, N = 3 SE +/- 379.42, N = 5 SE +/- 241.98, N = 3 25569 25552 25432 24965 25028 24788 25509 24967 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 24507 / Avg: 25569.25 / Max: 26559 Min: 24685 / Avg: 25552 / Max: 26239 Min: 24817 / Avg: 25432 / Max: 25925 Min: 24890 / Avg: 24965 / Max: 25013 Min: 24218 / Avg: 25028.33 / Max: 25573 Min: 24548 / Avg: 24788.33 / Max: 24953 Min: 24088 / Avg: 25509 / Max: 26344 Min: 24526 / Avg: 24967.33 / Max: 25360 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K SE +/- 950.23, N = 3 SE +/- 782.09, N = 3 SE +/- 743.92, N = 3 SE +/- 1806.17, N = 3 SE +/- 853.33, N = 3 SE +/- 512.00, N = 3 SE +/- 1288.51, N = 3 SE +/- 1458.18, N = 3 169045 173312 168021 169557 168533 169728 169216 169387 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K Min: 167168 / Avg: 169045.33 / Max: 170240 Min: 172288 / Avg: 173312 / Max: 174848 Min: 166656 / Avg: 168021.33 / Max: 169216 Min: 166144 / Avg: 169557.33 / Max: 172288 Min: 167680 / Avg: 168533.33 / Max: 170240 Min: 169216 / Avg: 169728 / Max: 170752 Min: 166656 / Avg: 169216 / Max: 170752 Min: 167680 / Avg: 169386.67 / Max: 172288 1. (CC) gcc options: -O2 -lpthread
O2 O3 OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Floating Point GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K 31348 31319 31066 30796 30742 31640 30677 30932 1. (CC) gcc options: -O3 -march=native -pipe
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 SE +/- 14.01, N = 4 SE +/- 15.47, N = 4 SE +/- 24.50, N = 4 SE +/- 7.08, N = 4 SE +/- 15.18, N = 4 SE +/- 3.97, N = 4 SE +/- 18.93, N = 4 SE +/- 11.51, N = 4 2494 2486 2473 2510 2493 2456 2533 2471 Result Confidence
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 2465 / Avg: 2493.5 / Max: 2530 Min: 2447 / Avg: 2486.25 / Max: 2512 Min: 2425 / Avg: 2472.5 / Max: 2541 Min: 2496 / Avg: 2510 / Max: 2525 Min: 2466 / Avg: 2492.75 / Max: 2521 Min: 2446 / Avg: 2456.25 / Max: 2463 Min: 2484 / Avg: 2533 / Max: 2576 Min: 2448 / Avg: 2470.75 / Max: 2501
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 14K 28K 42K 56K 70K SE +/- 733.11, N = 9 SE +/- 1101.74, N = 12 SE +/- 264.92, N = 3 SE +/- 1076.52, N = 15 SE +/- 52.81, N = 3 SE +/- 152.97, N = 3 SE +/- 663.99, N = 3 SE +/- 981.22, N = 5 64631 63921 65349 63921 65863 65915 63961 64401 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K Min: 58904.74 / Avg: 64630.82 / Max: 65928.41 Min: 51832.36 / Avg: 63920.7 / Max: 65431.31 Min: 64954.01 / Avg: 65348.55 / Max: 65852.09 Min: 52072.21 / Avg: 63920.84 / Max: 66073.04 Min: 65767.5 / Avg: 65863.47 / Max: 65949.64 Min: 65619.99 / Avg: 65914.97 / Max: 66132.78 Min: 62737.41 / Avg: 63960.55 / Max: 65020 Min: 60545.14 / Avg: 64400.99 / Max: 65979.38 1. (CC) gcc options: -O2
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite GRAPHITE LTO LTO GRAPHITE 160K 320K 480K 640K 800K SE +/- 1183.61, N = 3 SE +/- 591.64, N = 3 SE +/- 41.28, N = 3 SE +/- 1052.30, N = 3 SE +/- 4161.96, N = 3 SE +/- 1711.07, N = 3 SE +/- 402.40, N = 3 SE +/- 525.67, N = 3 737312 759979 738154 758399 754742 748357 758170 749663 Result Confidence
O2 O3 OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite GRAPHITE LTO LTO GRAPHITE 130K 260K 390K 520K 650K Min: 735058 / Avg: 737312.33 / Max: 739065 Min: 759285 / Avg: 759979 / Max: 761156 Min: 738083 / Avg: 738154 / Max: 738226 Min: 756434 / Avg: 758399.33 / Max: 760034 Min: 746659 / Avg: 754741.67 / Max: 760506 Min: 745016 / Avg: 748357 / Max: 750669 Min: 757708 / Avg: 758170.33 / Max: 758972 Min: 748778 / Avg: 749663 / Max: 750597
System Libxml2 Parsing This test measures the time to parse a random XML file with libxml2 via xmllint using the streaming API. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 850 KB GRAPHITE LTO LTO GRAPHITE 100 200 300 400 500 SE +/- 7.02, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 1.00, N = 3 SE +/- 4.00, N = 3 467 464 467 470 459 456 456 465 Result Confidence
O2 O3 OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 850 KB GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 Min: 459 / Avg: 467 / Max: 481 Min: 463 / Avg: 463.67 / Max: 464 Min: 466 / Avg: 466.67 / Max: 467 Min: 469 / Avg: 470.33 / Max: 472 Min: 458 / Avg: 459 / Max: 460 Min: 455 / Avg: 456 / Max: 457 Min: 455 / Avg: 456 / Max: 458 Min: 461 / Avg: 465 / Max: 473
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 90K 180K 270K 360K 450K SE +/- 5941.91, N = 3 SE +/- 1803.53, N = 3 SE +/- 6218.02, N = 12 SE +/- 5868.48, N = 15 SE +/- 1460.80, N = 3 SE +/- 3395.46, N = 3 SE +/- 619.63, N = 3 SE +/- 3509.18, N = 3 414131 417010 412802 404674 414306 411191 411027 409933 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 70K 140K 210K 280K 350K Min: 403785.49 / Avg: 414130.99 / Max: 424368.01 Min: 413403.31 / Avg: 417010.37 / Max: 418813.91 Min: 349130.58 / Avg: 412802.31 / Max: 430613.96 Min: 348892.68 / Avg: 404673.63 / Max: 428272.69 Min: 411410.21 / Avg: 414305.68 / Max: 416091.02 Min: 404743.08 / Avg: 411191.25 / Max: 416260.16 Min: 409927.94 / Avg: 411026.76 / Max: 412072.44 Min: 403626.33 / Avg: 409933.44 / Max: 415753.15 1. (CC) gcc options: -O2
Apache Siege This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 148.69, N = 3 SE +/- 479.40, N = 3 SE +/- 309.42, N = 3 SE +/- 314.91, N = 3 SE +/- 245.85, N = 3 SE +/- 386.76, N = 3 SE +/- 212.62, N = 3 SE +/- 151.94, N = 3 34820 34354 34648 34808 34336 34326 33794 33816 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 34626.04 / Avg: 34820.21 / Max: 35112.36 Min: 33692.72 / Avg: 34353.9 / Max: 35285.82 Min: 34270.05 / Avg: 34647.51 / Max: 35260.93 Min: 34270.05 / Avg: 34808.48 / Max: 35360.68 Min: 34013.61 / Avg: 34336.3 / Max: 34818.94 Min: 33829.5 / Avg: 34325.72 / Max: 35087.72 Min: 33400.13 / Avg: 33794.08 / Max: 34129.7 Min: 33557.05 / Avg: 33815.61 / Max: 34083.16 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 SE +/- 29.48, N = 3 SE +/- 12.23, N = 3 SE +/- 43.15, N = 3 SE +/- 46.75, N = 3 SE +/- 20.22, N = 3 SE +/- 52.15, N = 3 SE +/- 46.53, N = 3 SE +/- 27.45, N = 3 7346 7211 7296 7205 7258 7384 7227 7167 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 Min: 7288.04 / Avg: 7345.91 / Max: 7384.58 Min: 7189.86 / Avg: 7210.83 / Max: 7232.23 Min: 7222.78 / Avg: 7296.08 / Max: 7372.17 Min: 7111.85 / Avg: 7204.79 / Max: 7260.08 Min: 7227.49 / Avg: 7257.73 / Max: 7296.1 Min: 7324.32 / Avg: 7383.93 / Max: 7487.86 Min: 7155.42 / Avg: 7227.33 / Max: 7314.44 Min: 7113.93 / Avg: 7166.68 / Max: 7206.23 1. (CC) gcc options: -O3
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 10.97, N = 3 SE +/- 6.12, N = 3 SE +/- 3.90, N = 3 SE +/- 8.35, N = 3 SE +/- 5.59, N = 3 SE +/- 3.17, N = 3 SE +/- 11.69, N = 3 SE +/- 4.43, N = 3 3739 3751 3829 3725 3750 3837 3812 3748 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 3717.14 / Avg: 3738.58 / Max: 3753.35 Min: 3739.29 / Avg: 3751.36 / Max: 3759.18 Min: 3821.48 / Avg: 3829.24 / Max: 3833.79 Min: 3713.02 / Avg: 3725.3 / Max: 3741.25 Min: 3742.31 / Avg: 3749.84 / Max: 3760.77 Min: 3832.24 / Avg: 3837.45 / Max: 3843.19 Min: 3793.82 / Avg: 3811.66 / Max: 3833.66 Min: 3740.05 / Avg: 3747.66 / Max: 3755.41 1. (CC) gcc options: -lm -ldl
Unpacking The Linux Kernel This test measures how long it takes to extract the .tar.xz Linux kernel package. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Unpacking The Linux Kernel linux-4.15.tar.xz GRAPHITE LTO LTO GRAPHITE 1.1723 2.3446 3.5169 4.6892 5.8615 SE +/- 0.00, N = 4 SE +/- 0.01, N = 4 SE +/- 0.01, N = 4 SE +/- 0.01, N = 4 SE +/- 0.01, N = 4 SE +/- 0.02, N = 4 SE +/- 0.01, N = 4 SE +/- 0.01, N = 4 5.21 5.15 5.12 5.16 5.06 5.08 5.12 5.16 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Unpacking The Linux Kernel linux-4.15.tar.xz GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 5.2 / Avg: 5.21 / Max: 5.22 Min: 5.14 / Avg: 5.15 / Max: 5.17 Min: 5.1 / Avg: 5.12 / Max: 5.14 Min: 5.15 / Avg: 5.16 / Max: 5.18 Min: 5.05 / Avg: 5.06 / Max: 5.08 Min: 5.06 / Avg: 5.08 / Max: 5.14 Min: 5.1 / Avg: 5.12 / Max: 5.14 Min: 5.14 / Avg: 5.16 / Max: 5.18
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 0.54, N = 3 SE +/- 0.06, N = 3 SE +/- 0.50, N = 3 SE +/- 0.57, N = 3 SE +/- 0.72, N = 3 SE +/- 0.57, N = 3 SE +/- 0.60, N = 3 SE +/- 1.08, N = 3 172 172 169 174 170 174 173 174 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 171.5 / Avg: 172.49 / Max: 173.33 Min: 171.61 / Avg: 171.71 / Max: 171.8 Min: 168.44 / Avg: 169.13 / Max: 170.1 Min: 172.99 / Avg: 174.11 / Max: 174.84 Min: 168.57 / Avg: 170 / Max: 170.83 Min: 173.04 / Avg: 173.62 / Max: 174.77 Min: 171.67 / Avg: 172.61 / Max: 173.74 Min: 172.05 / Avg: 174.12 / Max: 175.67 1. (CC) gcc options: -O2
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K SE +/- 799.30, N = 3 SE +/- 756.06, N = 3 SE +/- 559.09, N = 3 SE +/- 1234.15, N = 3 SE +/- 141.80, N = 3 SE +/- 168.77, N = 3 SE +/- 516.25, N = 3 SE +/- 256.22, N = 3 61073 62586 61855 62351 60814 62497 61693 61102 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K Min: 59914 / Avg: 61073 / Max: 62606 Min: 61075 / Avg: 62586 / Max: 63392 Min: 61208 / Avg: 61854.67 / Max: 62968 Min: 60363 / Avg: 62351 / Max: 64612 Min: 60594 / Avg: 60814 / Max: 61079 Min: 62170 / Avg: 62497 / Max: 62733 Min: 60809 / Avg: 61692.67 / Max: 62597 Min: 60761 / Avg: 61102.33 / Max: 61604 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1600 3200 4800 6400 8000 SE +/- 18.21, N = 3 SE +/- 12.03, N = 3 SE +/- 39.04, N = 3 SE +/- 32.27, N = 3 SE +/- 50.82, N = 3 SE +/- 58.15, N = 3 SE +/- 38.21, N = 3 SE +/- 25.86, N = 3 7163 7046 7180 7038 7070 7240 7243 7072 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Read Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 Min: 7128.59 / Avg: 7162.93 / Max: 7190.63 Min: 7021.83 / Avg: 7045.88 / Max: 7058.44 Min: 7126.92 / Avg: 7179.55 / Max: 7255.83 Min: 6993.42 / Avg: 7037.52 / Max: 7100.39 Min: 7016.75 / Avg: 7070.25 / Max: 7171.85 Min: 7175.57 / Avg: 7240.32 / Max: 7356.35 Min: 7195.81 / Avg: 7242.64 / Max: 7318.36 Min: 7028.41 / Avg: 7071.61 / Max: 7117.83 1. (CC) gcc options: -O3
LZMA Compression This test measures the time needed to compress a file using LZMA compression. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better LZMA Compression 256MB File Compression GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 SE +/- 0.17, N = 3 SE +/- 0.25, N = 3 SE +/- 0.44, N = 3 SE +/- 0.60, N = 3 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 SE +/- 0.06, N = 3 SE +/- 0.20, N = 3 214 210 209 208 210 214 214 214 1. (CXX) g++ options: -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better LZMA Compression 256MB File Compression GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 Min: 213.74 / Avg: 214.01 / Max: 214.31 Min: 210 / Avg: 210.33 / Max: 210.83 Min: 208.12 / Avg: 208.74 / Max: 209.6 Min: 207.48 / Avg: 208.12 / Max: 209.32 Min: 209.5 / Avg: 209.67 / Max: 209.86 Min: 213.71 / Avg: 214.09 / Max: 214.32 Min: 213.96 / Avg: 214.02 / Max: 214.15 Min: 214 / Avg: 214.23 / Max: 214.63 1. (CXX) g++ options: -march=native -pipe -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 71.36, N = 3 SE +/- 60.10, N = 3 SE +/- 56.10, N = 3 SE +/- 61.55, N = 3 SE +/- 110.88, N = 3 SE +/- 45.45, N = 3 SE +/- 89.34, N = 3 SE +/- 39.48, N = 3 20709 20728 20988 20404 20573 20600 20728 20517 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 20583 / Avg: 20709.33 / Max: 20830 Min: 20662 / Avg: 20728 / Max: 20848 Min: 20883 / Avg: 20987.67 / Max: 21075 Min: 20287 / Avg: 20403.67 / Max: 20496 Min: 20352 / Avg: 20572.67 / Max: 20702 Min: 20510 / Avg: 20600 / Max: 20656 Min: 20591 / Avg: 20728.33 / Max: 20896 Min: 20441 / Avg: 20517.33 / Max: 20573 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 128 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 6.92, N = 3 SE +/- 5.87, N = 3 SE +/- 5.85, N = 3 SE +/- 6.10, N = 3 SE +/- 7.14, N = 3 SE +/- 232.93, N = 3 SE +/- 15.40, N = 3 SE +/- 5.91, N = 3 12512 12264 12316 12563 12502 12261 12538 12608 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 128 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 12501.7 / Avg: 12511.73 / Max: 12525 Min: 12254.7 / Avg: 12264.47 / Max: 12275 Min: 12309.7 / Avg: 12315.7 / Max: 12327.4 Min: 12553.9 / Avg: 12562.83 / Max: 12574.5 Min: 12488.3 / Avg: 12501.77 / Max: 12512.6 Min: 11795.3 / Avg: 12261.17 / Max: 12495.3 Min: 12510 / Avg: 12538.27 / Max: 12563 Min: 12599.3 / Avg: 12607.87 / Max: 12619.2 1. (CC) gcc options: -lpopt -march=native -pipe -O3
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 1.16, N = 3 SE +/- 1.17, N = 3 SE +/- 0.60, N = 3 SE +/- 4.53, N = 3 SE +/- 3.79, N = 3 SE +/- 2.82, N = 3 SE +/- 1.30, N = 3 SE +/- 0.50, N = 3 1232 1247 1242 1217 1225 1251 1241 1226 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1230.85 / Avg: 1232.35 / Max: 1234.64 Min: 1244.54 / Avg: 1246.77 / Max: 1248.48 Min: 1241.76 / Avg: 1242.4 / Max: 1243.6 Min: 1210.62 / Avg: 1216.76 / Max: 1225.61 Min: 1220.57 / Avg: 1225.47 / Max: 1232.92 Min: 1245.18 / Avg: 1250.78 / Max: 1254.14 Min: 1239.32 / Avg: 1241.42 / Max: 1243.8 Min: 1225.14 / Avg: 1226.01 / Max: 1226.86 1. (CC) gcc options: -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 1230.69, N = 3 SE +/- 1806.17, N = 3 SE +/- 1536.00, N = 3 SE +/- 1487.84, N = 3 SE +/- 615.35, N = 3 SE +/- 1038.12, N = 3 SE +/- 1396.97, N = 3 137643 136107 138496 137643 136960 137301 134741 136277 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 135936 / Avg: 137642.67 / Max: 140032 Min: 133376 / Avg: 136106.67 / Max: 139520 Min: 135424 / Avg: 138496 / Max: 140032 Min: 134912 / Avg: 137642.67 / Max: 140032 Min: 136448 / Avg: 137301.33 / Max: 138496 Min: 132864 / Avg: 134741.33 / Max: 136448 Min: 134400 / Avg: 136277.33 / Max: 139008 1. (CC) gcc options: -O2 -lpthread
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 72.59, N = 3 SE +/- 86.85, N = 3 SE +/- 147.79, N = 3 SE +/- 73.63, N = 3 SE +/- 103.49, N = 3 SE +/- 85.43, N = 3 SE +/- 140.33, N = 3 SE +/- 137.69, N = 3 25043 25458 24993 25060 25093 24771 24939 24872 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 24899 / Avg: 25043.33 / Max: 25129 Min: 25284 / Avg: 25457.67 / Max: 25547 Min: 24757 / Avg: 24992.67 / Max: 25265 Min: 24940 / Avg: 25060.33 / Max: 25194 Min: 24905 / Avg: 25093 / Max: 25262 Min: 24603 / Avg: 24771 / Max: 24882 Min: 24693 / Avg: 24939 / Max: 25179 Min: 24722 / Avg: 24872 / Max: 25147 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.00, N = 3 SE +/- 1.76, N = 3 SE +/- 1.67, N = 3 SE +/- 2.19, N = 3 SE +/- 1.00, N = 3 SE +/- 1.00, N = 3 SE +/- 1.15, N = 3 111 109 109 110 112 110 109 111 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 110 / Avg: 111 / Max: 113 Min: 106 / Avg: 109.33 / Max: 112 Min: 107 / Avg: 108.67 / Max: 112 Min: 107 / Avg: 109.67 / Max: 114 Min: 110 / Avg: 112 / Max: 113 Min: 108 / Avg: 109 / Max: 111 Min: 109 / Avg: 111 / Max: 113 1. (CC) gcc options: -O2 -lpthread
TSCP This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K SE +/- 889.16, N = 5 SE +/- 889.16, N = 5 SE +/- 1427.82, N = 5 SE +/- 1705.90, N = 5 SE +/- 726.00, N = 5 SE +/- 889.16, N = 5 SE +/- 765.00, N = 5 SE +/- 1206.25, N = 5 1414429 1414429 1450184 1449423 1415155 1413703 1452475 1449419 1. (CC) gcc options: -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K Min: 1412251 / Avg: 1414429 / Max: 1415881 Min: 1412251 / Avg: 1414429 / Max: 1415881 Min: 1445611 / Avg: 1450184.2 / Max: 1453240 Min: 1445611 / Avg: 1449423.4 / Max: 1453240 Min: 1412251 / Avg: 1415155 / Max: 1415881 Min: 1412251 / Avg: 1413703 / Max: 1415881 Min: 1449415 / Avg: 1452475 / Max: 1453240 Min: 1445611 / Avg: 1449419.2 / Max: 1453240 1. (CC) gcc options: -march=native -pipe -O3
Multichase Pointer Chaser This is a benchmark of Google's multichase pointer chaser program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 4MB Array, 64 Byte Stride GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 8.52 8.46 8.50 8.63 8.69 8.60 8.53 8.55 1. (CC) gcc options: -O2 -static -pthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 4MB Array, 64 Byte Stride GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 8.44 / Avg: 8.52 / Max: 8.67 Min: 8.43 / Avg: 8.46 / Max: 8.51 Min: 8.41 / Avg: 8.5 / Max: 8.68 Min: 8.57 / Avg: 8.63 / Max: 8.72 Min: 8.6 / Avg: 8.69 / Max: 8.81 Min: 8.52 / Avg: 8.6 / Max: 8.66 Min: 8.4 / Avg: 8.53 / Max: 8.66 Min: 8.47 / Avg: 8.55 / Max: 8.6 1. (CC) gcc options: -O2 -static -pthread -lrt
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 426.67, N = 3 SE +/- 938.67, N = 3 SE +/- 1194.67, N = 3 SE +/- 903.08, N = 3 SE +/- 225.77, N = 3 SE +/- 559.57, N = 3 SE +/- 853.33, N = 3 SE +/- 743.92, N = 3 96981 95275 96811 94421 94933 95531 94421 96981 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 96128 / Avg: 96981.33 / Max: 97408 Min: 94336 / Avg: 95274.67 / Max: 97152 Min: 94592 / Avg: 96810.67 / Max: 98688 Min: 93056 / Avg: 94421.33 / Max: 96128 Min: 94592 / Avg: 94933.33 / Max: 95360 Min: 94848 / Avg: 95530.67 / Max: 96640 Min: 93568 / Avg: 94421.33 / Max: 96128 Min: 95616 / Avg: 96981.33 / Max: 98176 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 85.33, N = 3 SE +/- 112.89, N = 3 SE +/- 586.57, N = 3 SE +/- 195.52, N = 3 SE +/- 279.78, N = 3 SE +/- 259.53, N = 3 SE +/- 225.77, N = 3 SE +/- 298.67, N = 3 39531 39957 40256 40128 40469 40597 39787 40085 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 39360 / Avg: 39530.67 / Max: 39616 Min: 39744 / Avg: 39957.33 / Max: 40128 Min: 39104 / Avg: 40256 / Max: 41024 Min: 39744 / Avg: 40128 / Max: 40384 Min: 40128 / Avg: 40469.33 / Max: 41024 Min: 40128 / Avg: 40597.33 / Max: 41024 Min: 39360 / Avg: 39786.67 / Max: 40128 Min: 39616 / Avg: 40085.33 / Max: 40640 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 50K 100K 150K 200K 250K SE +/- 3105.01, N = 3 SE +/- 1645.85, N = 3 SE +/- 1923.32, N = 3 SE +/- 3426.11, N = 3 SE +/- 615.35, N = 3 SE +/- 1487.84, N = 3 SE +/- 2960.96, N = 3 SE +/- 2048.00, N = 3 223659 219904 219563 221611 219051 222805 218027 217856 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K Min: 218368 / Avg: 223658.67 / Max: 229120 Min: 216832 / Avg: 219904 / Max: 222464 Min: 216320 / Avg: 219562.67 / Max: 222976 Min: 214784 / Avg: 221610.67 / Max: 225536 Min: 217856 / Avg: 219050.67 / Max: 219904 Min: 220416 / Avg: 222805.33 / Max: 225536 Min: 212224 / Avg: 218026.67 / Max: 221952 Min: 215808 / Avg: 217856 / Max: 221952 1. (CC) gcc options: -O2 -lpthread
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil GRAPHITE LTO LTO GRAPHITE 1.305 2.61 3.915 5.22 6.525 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 5 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 5.77 5.80 5.79 5.70 5.77 5.65 5.73 5.77 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 5.67 / Avg: 5.77 / Max: 5.85 Min: 5.68 / Avg: 5.8 / Max: 5.89 Min: 5.77 / Avg: 5.79 / Max: 5.81 Min: 5.67 / Avg: 5.7 / Max: 5.77 Min: 5.44 / Avg: 5.77 / Max: 5.92 Min: 5.54 / Avg: 5.65 / Max: 5.7 Min: 5.51 / Avg: 5.73 / Max: 5.88 Min: 5.75 / Avg: 5.77 / Max: 5.8 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
asmFish This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth GRAPHITE LTO LTO GRAPHITE 14M 28M 42M 56M 70M SE +/- 524774.43, N = 3 SE +/- 224301.53, N = 3 SE +/- 1212462.43, N = 3 SE +/- 83485.52, N = 3 SE +/- 34223.87, N = 3 SE +/- 183322.24, N = 3 SE +/- 522758.71, N = 3 SE +/- 859562.02, N = 3 63379126 64287169 63389475 62631352 63414993 64033269 63600949 62693072 Result Confidence
O2 O3 OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth GRAPHITE LTO LTO GRAPHITE 11M 22M 33M 44M 55M Min: 62366264 / Avg: 63379125.67 / Max: 64123775 Min: 63983110 / Avg: 64287168.67 / Max: 64724846 Min: 61134577 / Avg: 63389475 / Max: 65289434 Min: 62523965 / Avg: 62631352.33 / Max: 62795773 Min: 63348507 / Avg: 63414993 / Max: 63462326 Min: 63771300 / Avg: 64033269.33 / Max: 64386403 Min: 63028689 / Avg: 63600948.67 / Max: 64644851 Min: 61504188 / Avg: 62693072.33 / Max: 64362901
XZ Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using XZ compression. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 SE +/- 0.12, N = 3 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 17.05 16.91 16.83 16.96 17.02 17.16 16.82 16.72 1. (CC) gcc options: -pthread -fvisibility=hidden -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 17.01 / Avg: 17.05 / Max: 17.09 Min: 16.82 / Avg: 16.91 / Max: 16.97 Min: 16.77 / Avg: 16.83 / Max: 16.9 Min: 16.84 / Avg: 16.96 / Max: 17.14 Min: 16.84 / Avg: 17.02 / Max: 17.33 Min: 16.92 / Avg: 17.16 / Max: 17.29 Min: 16.64 / Avg: 16.82 / Max: 17.16 Min: 16.46 / Avg: 16.72 / Max: 16.92 1. (CC) gcc options: -pthread -fvisibility=hidden -march=native -pipe -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 SE +/- 3.87, N = 3 SE +/- 1.12, N = 3 SE +/- 1.40, N = 3 SE +/- 2.30, N = 3 SE +/- 6.19, N = 3 SE +/- 1.80, N = 3 SE +/- 0.59, N = 3 SE +/- 3.06, N = 3 380 390 383 390 386 390 385 384 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 372.41 / Avg: 380.14 / Max: 384.04 Min: 388.33 / Avg: 390.27 / Max: 392.22 Min: 379.79 / Avg: 382.58 / Max: 384.25 Min: 385.51 / Avg: 389.76 / Max: 393.43 Min: 374.49 / Avg: 385.71 / Max: 395.85 Min: 386.19 / Avg: 389.78 / Max: 391.69 Min: 383.78 / Avg: 384.82 / Max: 385.81 Min: 378.26 / Avg: 383.97 / Max: 388.75 1. (CC) gcc options: -O2
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 256 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 9.93, N = 3 SE +/- 19.39, N = 3 SE +/- 13.24, N = 3 SE +/- 28.96, N = 3 SE +/- 22.21, N = 3 SE +/- 2.76, N = 3 SE +/- 8.74, N = 3 SE +/- 52.91, N = 3 12400 12102 12142 12391 12363 12384 12419 12398 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 256 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 12383.5 / Avg: 12399.57 / Max: 12417.7 Min: 12065.6 / Avg: 12101.6 / Max: 12132.1 Min: 12118 / Avg: 12142 / Max: 12163.7 Min: 12337.4 / Avg: 12391.03 / Max: 12436.8 Min: 12319.1 / Avg: 12362.5 / Max: 12392.4 Min: 12380.1 / Avg: 12383.97 / Max: 12389.3 Min: 12401.9 / Avg: 12418.73 / Max: 12431.2 Min: 12294.7 / Avg: 12398.43 / Max: 12468.4 1. (CC) gcc options: -lpopt -march=native -pipe -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 677.31, N = 3 SE +/- 1089.46, N = 3 SE +/- 1373.31, N = 3 SE +/- 1353.17, N = 5 SE +/- 950.23, N = 3 SE +/- 170.67, N = 3 SE +/- 1115.88, N = 3 SE +/- 1480.48, N = 3 89728 88875 90411 88755 90155 88107 88192 89557 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K Min: 88448 / Avg: 89728 / Max: 90752 Min: 87424 / Avg: 88874.67 / Max: 91008 Min: 88448 / Avg: 90410.67 / Max: 93056 Min: 84096 / Avg: 88755.2 / Max: 91776 Min: 88960 / Avg: 90154.67 / Max: 92032 Min: 87936 / Avg: 88106.67 / Max: 88448 Min: 86400 / Avg: 88192 / Max: 90240 Min: 86912 / Avg: 89557.33 / Max: 92032 1. (CC) gcc options: -O2 -lpthread
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 603.41, N = 3 SE +/- 694.07, N = 3 SE +/- 629.85, N = 3 SE +/- 604.79, N = 3 SE +/- 1026.94, N = 3 SE +/- 31.19, N = 3 SE +/- 635.46, N = 3 SE +/- 310.91, N = 3 55386 56445 55849 55134 55337 56569 55845 55622 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 54184 / Avg: 55385.67 / Max: 56083 Min: 55080 / Avg: 56445.33 / Max: 57345 Min: 54607 / Avg: 55849.33 / Max: 56651 Min: 54082 / Avg: 55134 / Max: 56177 Min: 53504 / Avg: 55336.67 / Max: 57056 Min: 56514 / Avg: 56569 / Max: 56622 Min: 54587 / Avg: 55844.67 / Max: 56632 Min: 55070 / Avg: 55621.67 / Max: 56146 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K SE +/- 192.25, N = 3 SE +/- 221.71, N = 3 SE +/- 67.37, N = 3 SE +/- 144.54, N = 3 SE +/- 159.52, N = 3 SE +/- 109.13, N = 3 SE +/- 200.46, N = 3 SE +/- 246.03, N = 3 29711 29699 29780 29294 29516 29883 30041 29757 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K Min: 29430.36 / Avg: 29710.62 / Max: 30078.72 Min: 29255.47 / Avg: 29698.84 / Max: 29925.77 Min: 29657.09 / Avg: 29780.44 / Max: 29889.08 Min: 29024.94 / Avg: 29293.72 / Max: 29520.3 Min: 29275.55 / Avg: 29515.78 / Max: 29817.72 Min: 29731.14 / Avg: 29882.9 / Max: 30094.63 Min: 29641.64 / Avg: 30040.83 / Max: 30272.57 Min: 29435.44 / Avg: 29757.07 / Max: 30240.39 1. (CC) gcc options: -O2
PostMark This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.
Dbench Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 1 GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 4.84, N = 3 SE +/- 0.50, N = 3 SE +/- 1.09, N = 3 SE +/- 3.69, N = 3 SE +/- 3.34, N = 3 SE +/- 6.70, N = 3 SE +/- 2.96, N = 3 SE +/- 2.60, N = 3 1041 1040 1065 1061 1048 1041 1061 1066 1. (CC) gcc options: -lpopt -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 1 GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1031.59 / Avg: 1041.19 / Max: 1047.02 Min: 1038.86 / Avg: 1039.55 / Max: 1040.51 Min: 1062.78 / Avg: 1064.81 / Max: 1066.51 Min: 1055.24 / Avg: 1061.03 / Max: 1067.88 Min: 1041.49 / Avg: 1047.98 / Max: 1052.57 Min: 1031.02 / Avg: 1040.99 / Max: 1053.72 Min: 1057.07 / Avg: 1061.21 / Max: 1066.94 Min: 1061.91 / Avg: 1066.15 / Max: 1070.88 1. (CC) gcc options: -lpopt -march=native -pipe -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 658.27, N = 15 SE +/- 991.39, N = 3 SE +/- 841.58, N = 3 SE +/- 929.65, N = 3 SE +/- 697.74, N = 15 SE +/- 507.47, N = 11 SE +/- 579.02, N = 3 SE +/- 458.91, N = 15 50491 49264 50357 50321 49798 49594 50007 49600 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 512 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 46391 / Avg: 50490.53 / Max: 54491 Min: 48183 / Avg: 49264 / Max: 51244 Min: 48856 / Avg: 50357.33 / Max: 51767 Min: 48545 / Avg: 50320.67 / Max: 51686 Min: 46552 / Avg: 49798.27 / Max: 54043 Min: 47956 / Avg: 49593.64 / Max: 53280 Min: 49021 / Avg: 50007.33 / Max: 51026 Min: 45726 / Avg: 49599.73 / Max: 54016 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Thread GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 SE +/- 0.16, N = 3 SE +/- 0.31, N = 3 SE +/- 0.17, N = 3 SE +/- 0.24, N = 3 SE +/- 0.21, N = 3 23.39 23.90 23.63 23.60 23.97 23.65 23.48 23.66 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Thread GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 Min: 23.3 / Avg: 23.39 / Max: 23.5 Min: 23.59 / Avg: 23.9 / Max: 24.11 Min: 23.35 / Avg: 23.63 / Max: 23.77 Min: 23.29 / Avg: 23.6 / Max: 23.84 Min: 23.41 / Avg: 23.97 / Max: 24.48 Min: 23.31 / Avg: 23.65 / Max: 23.85 Min: 23.23 / Avg: 23.48 / Max: 23.95 Min: 23.28 / Avg: 23.66 / Max: 23.99 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Append GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 406.38, N = 3 SE +/- 197.87, N = 3 SE +/- 481.97, N = 3 SE +/- 1047.00, N = 3 SE +/- 300.12, N = 3 SE +/- 152.54, N = 3 SE +/- 180.97, N = 3 SE +/- 248.84, N = 3 78453 79119 77213 78950 78778 78707 77782 78750 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Append GRAPHITE LTO LTO GRAPHITE 14K 28K 42K 56K 70K Min: 77837.5 / Avg: 78453.33 / Max: 79220.6 Min: 78854.1 / Avg: 79119.33 / Max: 79506.3 Min: 76709 / Avg: 77212.8 / Max: 78176.4 Min: 77420.8 / Avg: 78949.73 / Max: 80953.3 Min: 78278 / Avg: 78778.03 / Max: 79315.6 Min: 78405.2 / Avg: 78707.47 / Max: 78894.4 Min: 77471.9 / Avg: 77782.27 / Max: 78098.7 Min: 78368.3 / Avg: 78749.5 / Max: 79217.2 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.22, N = 15 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.20, N = 3 12.21 12.30 12.27 12.48 12.20 12.22 12.18 12.22 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 12.16 / Avg: 12.21 / Max: 12.29 Min: 12.24 / Avg: 12.3 / Max: 12.4 Min: 12.1 / Avg: 12.27 / Max: 12.38 Min: 11.94 / Avg: 12.48 / Max: 14.74 Min: 12.09 / Avg: 12.2 / Max: 12.32 Min: 12.15 / Avg: 12.22 / Max: 12.28 Min: 12.09 / Avg: 12.18 / Max: 12.23 Min: 12.01 / Avg: 12.22 / Max: 12.61 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh GRAPHITE LTO LTO GRAPHITE 0.189 0.378 0.567 0.756 0.945 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.84 0.83 0.82 0.82 0.84 0.84 0.83 0.82 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 0.84 / Avg: 0.84 / Max: 0.85 Min: 0.83 / Avg: 0.83 / Max: 0.84 Min: 0.82 / Avg: 0.82 / Max: 0.82 Min: 0.82 / Avg: 0.82 / Max: 0.82 Min: 0.83 / Avg: 0.84 / Max: 0.84 Min: 0.84 / Avg: 0.84 / Max: 0.84 Min: 0.82 / Avg: 0.83 / Max: 0.83 Min: 0.82 / Avg: 0.82 / Max: 0.82 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Replace GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 212.11, N = 3 SE +/- 642.36, N = 3 SE +/- 287.02, N = 3 SE +/- 563.61, N = 3 SE +/- 385.17, N = 3 SE +/- 249.10, N = 3 SE +/- 297.20, N = 3 SE +/- 258.88, N = 3 78645 79352 78935 80017 79380 78940 78114 78980 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Replace GRAPHITE LTO LTO GRAPHITE 14K 28K 42K 56K 70K Min: 78261 / Avg: 78645.4 / Max: 78993 Min: 78260.4 / Avg: 79352.1 / Max: 80484.5 Min: 78482.4 / Avg: 78934.53 / Max: 79466.9 Min: 78895 / Avg: 80016.5 / Max: 80675.5 Min: 78674.8 / Avg: 79380.23 / Max: 80001 Min: 78471.8 / Avg: 78939.6 / Max: 79321.9 Min: 77569.8 / Avg: 78114.33 / Max: 78593 Min: 78720.6 / Avg: 78980.33 / Max: 79498.1 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 152.30, N = 3 SE +/- 52.33, N = 3 SE +/- 95.52, N = 3 SE +/- 65.69, N = 3 SE +/- 178.36, N = 3 SE +/- 67.60, N = 3 SE +/- 43.97, N = 3 SE +/- 110.01, N = 3 16169 16075 16291 16153 16025 16204 16411 16342 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 15867.11 / Avg: 16169.27 / Max: 16353.65 Min: 16022.03 / Avg: 16075.36 / Max: 16180 Min: 16112.79 / Avg: 16290.95 / Max: 16439.76 Min: 16022.03 / Avg: 16153.31 / Max: 16223.07 Min: 15668.99 / Avg: 16025.38 / Max: 16216.9 Min: 16070.31 / Avg: 16203.78 / Max: 16289.13 Min: 16351.56 / Avg: 16411.19 / Max: 16496.97 Min: 16124.97 / Avg: 16341.92 / Max: 16482.1 1. (CC) gcc options: -O2
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Total Time GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.12, N = 3 SE +/- 0.78, N = 3 SE +/- 0.62, N = 3 SE +/- 0.25, N = 3 SE +/- 0.54, N = 3 SE +/- 0.38, N = 3 SE +/- 0.71, N = 3 SE +/- 0.30, N = 3 129 130 130 128 127 127 127 127 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Total Time GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 127.46 / Avg: 128.98 / Max: 131.17 Min: 128.82 / Avg: 129.95 / Max: 131.45 Min: 128.82 / Avg: 129.51 / Max: 130.76 Min: 127.03 / Avg: 127.52 / Max: 127.85 Min: 126.04 / Avg: 127.08 / Max: 127.84 Min: 126.86 / Avg: 127.25 / Max: 128.01 Min: 126.17 / Avg: 127.38 / Max: 128.63 Min: 126.67 / Avg: 127.18 / Max: 127.7
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 299.93, N = 3 SE +/- 593.97, N = 3 SE +/- 306.40, N = 3 SE +/- 2231.84, N = 3 SE +/- 311.86, N = 3 SE +/- 956.22, N = 3 SE +/- 929.83, N = 3 SE +/- 503.64, N = 3 118859 119129 118726 116742 119458 117773 118725 118828 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 118415.73 / Avg: 118859.36 / Max: 119430.84 Min: 117945.17 / Avg: 119128.82 / Max: 119808.12 Min: 118210.68 / Avg: 118726.22 / Max: 119270.86 Min: 112293.02 / Avg: 116742.23 / Max: 119277.81 Min: 119000.58 / Avg: 119458.01 / Max: 120053.93 Min: 116172.22 / Avg: 117773.05 / Max: 119479.61 Min: 116995.14 / Avg: 118725.49 / Max: 120180.74 Min: 118067.57 / Avg: 118827.63 / Max: 119780.09 1. (CC) gcc options: -O2
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.C GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 42.81, N = 3 SE +/- 37.43, N = 3 SE +/- 53.04, N = 3 SE +/- 132.51, N = 3 SE +/- 21.73, N = 3 SE +/- 73.68, N = 3 SE +/- 49.13, N = 3 SE +/- 93.42, N = 3 43251 42702 43242 42592 43270 42526 43177 42299 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.C GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K Min: 43177.42 / Avg: 43251.11 / Max: 43325.71 Min: 42629.64 / Avg: 42702.36 / Max: 42754.14 Min: 43165.57 / Avg: 43242.48 / Max: 43344.2 Min: 42329.08 / Avg: 42591.64 / Max: 42754.13 Min: 43235.12 / Avg: 43270.45 / Max: 43310.02 Min: 42394.38 / Avg: 42525.67 / Max: 42649.26 Min: 43083.99 / Avg: 43177.49 / Max: 43250.4 Min: 42138.48 / Avg: 42298.7 / Max: 42462.05 1. 3.3
PolyBench-C PolyBench-C is a C-language polyhedral benchmark suite made at the Ohio State University. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: 3 Matrix Multiplications GRAPHITE LTO LTO GRAPHITE 0.6075 1.215 1.8225 2.43 3.0375 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 2.68 2.70 2.70 2.69 2.65 2.70 2.67 2.64 1. (CC) gcc options: -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: 3 Matrix Multiplications GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.68 / Avg: 2.68 / Max: 2.68 Min: 2.69 / Avg: 2.7 / Max: 2.71 Min: 2.69 / Avg: 2.7 / Max: 2.7 Min: 2.69 / Avg: 2.69 / Max: 2.69 Min: 2.65 / Avg: 2.65 / Max: 2.65 Min: 2.69 / Avg: 2.7 / Max: 2.7 Min: 2.64 / Avg: 2.67 / Max: 2.69 Min: 2.64 / Avg: 2.64 / Max: 2.65 1. (CC) gcc options: -march=native -pipe -O3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 23.35, N = 3 SE +/- 37.69, N = 3 SE +/- 42.85, N = 3 SE +/- 333.71, N = 12 SE +/- 46.86, N = 3 SE +/- 47.01, N = 3 SE +/- 15.25, N = 3 SE +/- 29.58, N = 3 16063 16025 16094 15784 16129 16130 16142 16125 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 16030.06 / Avg: 16063.15 / Max: 16108.23 Min: 15950.65 / Avg: 16025.38 / Max: 16071.32 Min: 16015.01 / Avg: 16094.28 / Max: 16162.13 Min: 12116.62 / Avg: 15784.26 / Max: 16188.7 Min: 16036.08 / Avg: 16128.63 / Max: 16187.68 Min: 16057.2 / Avg: 16129.98 / Max: 16217.93 Min: 16113.3 / Avg: 16142.11 / Max: 16165.19 Min: 16065.77 / Avg: 16124.91 / Max: 16156.01 1. (CC) gcc options: -O2
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 9.04 8.98 8.89 9.07 9.02 9.09 8.96 8.97 1. (CC) gcc options: -march=native -pipe -O3 -pthread -lz -llzma Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 8.99 / Avg: 9.04 / Max: 9.11 Min: 8.92 / Avg: 8.98 / Max: 9.04 Min: 8.84 / Avg: 8.89 / Max: 8.99 Min: 9.01 / Avg: 9.07 / Max: 9.15 Min: 8.98 / Avg: 9.02 / Max: 9.07 Min: 9.07 / Avg: 9.09 / Max: 9.13 Min: 8.94 / Avg: 8.96 / Max: 9 Min: 8.92 / Avg: 8.97 / Max: 9.04 1. (CC) gcc options: -march=native -pipe -O3 -pthread -lz -llzma
IOzone The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 3.29, N = 3 SE +/- 2.62, N = 3 SE +/- 1.63, N = 3 SE +/- 4.23, N = 3 SE +/- 4.40, N = 3 SE +/- 8.21, N = 3 SE +/- 9.91, N = 3 SE +/- 8.97, N = 3 1211 1230 1223 1221 1235 1227 1208 1220 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Write Performance GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1207.46 / Avg: 1211.31 / Max: 1217.85 Min: 1225.77 / Avg: 1230.07 / Max: 1234.82 Min: 1220.09 / Avg: 1223.22 / Max: 1225.57 Min: 1212.14 / Avg: 1220.58 / Max: 1225.4 Min: 1227.33 / Avg: 1235.07 / Max: 1242.55 Min: 1213.99 / Avg: 1227.2 / Max: 1242.26 Min: 1188.89 / Avg: 1208.04 / Max: 1222.04 Min: 1203.52 / Avg: 1220.01 / Max: 1234.38 1. (CC) gcc options: -O3
Multichase Pointer Chaser This is a benchmark of Google's multichase pointer chaser program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride GRAPHITE LTO LTO GRAPHITE 15 30 45 60 75 SE +/- 0.03, N = 3 SE +/- 0.23, N = 3 SE +/- 0.45, N = 3 SE +/- 0.07, N = 3 SE +/- 0.33, N = 3 SE +/- 0.18, N = 3 SE +/- 0.41, N = 3 SE +/- 0.37, N = 3 68.11 67.94 67.72 67.19 67.70 68.69 67.53 67.71 1. (CC) gcc options: -O2 -static -pthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 Min: 68.06 / Avg: 68.11 / Max: 68.17 Min: 67.5 / Avg: 67.94 / Max: 68.28 Min: 66.86 / Avg: 67.72 / Max: 68.38 Min: 67.04 / Avg: 67.19 / Max: 67.27 Min: 67.25 / Avg: 67.7 / Max: 68.34 Min: 68.45 / Avg: 68.69 / Max: 69.05 Min: 67.1 / Avg: 67.53 / Max: 68.35 Min: 66.99 / Avg: 67.71 / Max: 68.19 1. (CC) gcc options: -O2 -static -pthread -lrt
Sysbench This is a benchmark of Sysbench with CPU and memory sub-tests. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: Memory GRAPHITE LTO LTO GRAPHITE 4M 8M 12M 16M 20M SE +/- 9190.64, N = 3 SE +/- 157893.08, N = 3 SE +/- 101995.88, N = 3 SE +/- 143605.27, N = 3 SE +/- 73779.96, N = 3 SE +/- 83986.00, N = 3 SE +/- 22770.05, N = 3 SE +/- 41886.85, N = 3 19455798 19502832 19654284 19492222 19645196 19834659 19889024 19881909 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=native -pipe -rdynamic -ldl -laio -lm Result Confidence
O2 O3 OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: Memory GRAPHITE LTO LTO GRAPHITE 3M 6M 9M 12M 15M Min: 19443717.73 / Avg: 19455798.05 / Max: 19473836.25 Min: 19208990.33 / Avg: 19502832.28 / Max: 19749919.89 Min: 19513342.29 / Avg: 19654283.97 / Max: 19852469.56 Min: 19205139.04 / Avg: 19492222.05 / Max: 19643174.77 Min: 19510008.94 / Avg: 19645196.26 / Max: 19764012.35 Min: 19744246.68 / Avg: 19834659.45 / Max: 20002463.13 Min: 19843886.17 / Avg: 19889024.45 / Max: 19916821.11 Min: 19838494.97 / Avg: 19881908.79 / Max: 19965663.7 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=native -pipe -rdynamic -ldl -laio -lm
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 625.71, N = 3 SE +/- 459.66, N = 3 SE +/- 237.56, N = 3 SE +/- 42.67, N = 3 SE +/- 364.54, N = 3 SE +/- 348.75, N = 3 SE +/- 307.67, N = 3 SE +/- 213.33, N = 3 33248 33248 33643 33429 33259 33077 33813 33515 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 32032 / Avg: 33248 / Max: 34112 Min: 32544 / Avg: 33248 / Max: 34112 Min: 33344 / Avg: 33642.67 / Max: 34112 Min: 33344 / Avg: 33429.33 / Max: 33472 Min: 32832 / Avg: 33258.67 / Max: 33984 Min: 32416 / Avg: 33077.33 / Max: 33600 Min: 33216 / Avg: 33813.33 / Max: 34240 Min: 33088 / Avg: 33514.67 / Max: 33728 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K SE +/- 128.00, N = 3 SE +/- 279.78, N = 3 SE +/- 128.00, N = 3 SE +/- 364.54, N = 3 SE +/- 307.67, N = 3 SE +/- 112.89, N = 3 37952 37781 37824 37952 37141 37824 37227 37867 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 37824 / Avg: 37952 / Max: 38208 Min: 37440 / Avg: 37781.33 / Max: 38336 Min: 37696 / Avg: 37824 / Max: 38080 Min: 36416 / Avg: 37141.33 / Max: 37568 Min: 36800 / Avg: 37226.67 / Max: 37824 Min: 37696 / Avg: 37866.67 / Max: 38080 1. (CC) gcc options: -O2 -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K SE +/- 415.38, N = 3 SE +/- 576.21, N = 3 SE +/- 240.58, N = 3 SE +/- 52.53, N = 3 SE +/- 404.18, N = 3 SE +/- 250.53, N = 3 SE +/- 608.00, N = 3 SE +/- 234.56, N = 3 59513 59785 59244 60141 58867 59974 60071 59681 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 58725.7 / Avg: 59513.03 / Max: 60136.25 Min: 58638.26 / Avg: 59784.72 / Max: 60459.35 Min: 58783.01 / Avg: 59244.08 / Max: 59593.79 Min: 60083.32 / Avg: 60141.05 / Max: 60245.93 Min: 58105.88 / Avg: 58867.42 / Max: 59483.01 Min: 59531.42 / Avg: 59973.98 / Max: 60398.73 Min: 58945.43 / Avg: 60071.01 / Max: 61032.3 Min: 59214.71 / Avg: 59681.31 / Max: 59956.67 1. (CC) gcc options: -O2
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 118.78, N = 3 SE +/- 69.95, N = 3 SE +/- 28.22, N = 3 SE +/- 128.00, N = 3 SE +/- 21.33, N = 3 SE +/- 87.31, N = 3 SE +/- 73.90, N = 3 SE +/- 102.87, N = 3 12667 12635 12507 12656 12773 12635 12656 12688 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 6 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 12432 / Avg: 12666.67 / Max: 12816 Min: 12496 / Avg: 12634.67 / Max: 12720 Min: 12464 / Avg: 12506.67 / Max: 12560 Min: 12400 / Avg: 12656 / Max: 12784 Min: 12752 / Avg: 12773.33 / Max: 12816 Min: 12464 / Avg: 12634.67 / Max: 12752 Min: 12528 / Avg: 12656 / Max: 12784 Min: 12496 / Avg: 12688 / Max: 12848 1. (CC) gcc options: -O2 -lpthread
Multichase Pointer Chaser This is a benchmark of Google's multichase pointer chaser program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 256MB Array, 256 Byte Stride GRAPHITE LTO LTO GRAPHITE 14 28 42 56 70 SE +/- 0.48, N = 3 SE +/- 0.51, N = 3 SE +/- 0.23, N = 3 SE +/- 0.36, N = 3 SE +/- 0.07, N = 3 SE +/- 0.35, N = 3 SE +/- 0.51, N = 3 SE +/- 0.09, N = 3 62.70 62.02 61.88 61.40 62.23 61.75 62.12 62.60 1. (CC) gcc options: -O2 -static -pthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 256MB Array, 256 Byte Stride GRAPHITE LTO LTO GRAPHITE 12 24 36 48 60 Min: 61.92 / Avg: 62.7 / Max: 63.57 Min: 61.08 / Avg: 62.02 / Max: 62.85 Min: 61.61 / Avg: 61.88 / Max: 62.33 Min: 61 / Avg: 61.4 / Max: 62.13 Min: 62.16 / Avg: 62.23 / Max: 62.38 Min: 61.06 / Avg: 61.75 / Max: 62.14 Min: 61.18 / Avg: 62.12 / Max: 62.95 Min: 62.44 / Avg: 62.6 / Max: 62.73 1. (CC) gcc options: -O2 -static -pthread -lrt
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 758.46, N = 3 SE +/- 341.33, N = 3 SE +/- 890.91, N = 3 SE +/- 256.00, N = 3 SE +/- 903.08, N = 3 SE +/- 1239.54, N = 3 SE +/- 961.66, N = 3 98261 99285 99115 98944 99371 100053 98005 98688 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 97152 / Avg: 98261.33 / Max: 99712 Min: 98944 / Avg: 99285.33 / Max: 99968 Min: 97664 / Avg: 99114.67 / Max: 100736 Min: 98432 / Avg: 98944 / Max: 99200 Min: 97664 / Avg: 99370.67 / Max: 100736 Min: 98688 / Avg: 100053.33 / Max: 102528 Min: 96384 / Avg: 98005.33 / Max: 99712 1. (CC) gcc options: -O2 -lpthread
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Thread GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.16, N = 3 SE +/- 0.09, N = 3 SE +/- 0.17, N = 3 15.51 15.54 15.51 15.63 15.52 15.44 15.75 15.76 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Thread GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 15.41 / Avg: 15.51 / Max: 15.65 Min: 15.45 / Avg: 15.54 / Max: 15.69 Min: 15.4 / Avg: 15.51 / Max: 15.67 Min: 15.55 / Avg: 15.63 / Max: 15.73 Min: 15.32 / Avg: 15.52 / Max: 15.67 Min: 15.14 / Avg: 15.44 / Max: 15.65 Min: 15.58 / Avg: 15.75 / Max: 15.86 Min: 15.43 / Avg: 15.76 / Max: 16.01 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Read Compiled Tree GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 23.66, N = 3 SE +/- 8.94, N = 3 SE +/- 24.09, N = 3 SE +/- 17.63, N = 3 SE +/- 8.94, N = 3 SE +/- 35.54, N = 3 SE +/- 0.00, N = 3 SE +/- 9.50, N = 3 2701 2692 2728 2674 2674 2702 2683 2729 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Read Compiled Tree GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 Min: 2655.91 / Avg: 2700.62 / Max: 2736.39 Min: 2682.73 / Avg: 2691.67 / Max: 2709.56 Min: 2682.73 / Avg: 2728 / Max: 2764.89 Min: 2655.91 / Avg: 2674.3 / Max: 2709.56 Min: 2655.91 / Avg: 2673.79 / Max: 2682.73 Min: 2630.61 / Avg: 2701.69 / Max: 2738.07 Min: 2682.73 / Avg: 2682.73 / Max: 2682.73 Min: 2709.56 / Avg: 2728.57 / Max: 2738.07
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 174.62, N = 3 SE +/- 42.67, N = 3 SE +/- 10.67, N = 3 SE +/- 245.33, N = 3 SE +/- 10.67, N = 3 SE +/- 10.67, N = 3 SE +/- 10.67, N = 3 SE +/- 10.67, N = 3 13275 13051 13019 13285 13019 13019 13019 13029 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 13040 / Avg: 13274.67 / Max: 13616 Min: 13008 / Avg: 13050.67 / Max: 13136 Min: 13008 / Avg: 13018.67 / Max: 13040 Min: 13040 / Avg: 13285.33 / Max: 13776 Min: 13008 / Avg: 13018.67 / Max: 13040 Min: 13008 / Avg: 13018.67 / Max: 13040 Min: 13008 / Avg: 13018.67 / Max: 13040 Min: 13008 / Avg: 13029.33 / Max: 13040 1. (CC) gcc options: -O2 -lpthread
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Merge Results GRAPHITE LTO LTO GRAPHITE 0.684 1.368 2.052 2.736 3.42 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.02, N = 2 SE +/- 0.01, N = 2 SE +/- 0.04, N = 2 SE +/- 0.00, N = 2 SE +/- 0.02, N = 2 SE +/- 0.00, N = 2 3.00 3.03 3.04 3.02 3.03 2.99 3.00 2.98 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Merge Results GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.99 / Avg: 3 / Max: 3 Min: 3.03 / Avg: 3.03 / Max: 3.03 Min: 3.02 / Avg: 3.04 / Max: 3.06 Min: 3.02 / Avg: 3.02 / Max: 3.03 Min: 2.99 / Avg: 3.03 / Max: 3.07 Min: 2.99 / Avg: 2.99 / Max: 3 Min: 2.98 / Avg: 3 / Max: 3.03 Min: 2.98 / Avg: 2.98 / Max: 2.98
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 1.18, N = 3 SE +/- 0.87, N = 3 SE +/- 0.46, N = 3 SE +/- 1.23, N = 3 SE +/- 1.31, N = 3 SE +/- 3.57, N = 3 SE +/- 1.16, N = 3 SE +/- 1.13, N = 3 205 204 203 203 202 201 205 202 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 Min: 202.38 / Avg: 204.74 / Max: 205.95 Min: 202.62 / Avg: 204.35 / Max: 205.39 Min: 201.95 / Avg: 202.78 / Max: 203.53 Min: 200.62 / Avg: 202.98 / Max: 204.78 Min: 199.24 / Avg: 201.71 / Max: 203.71 Min: 194.53 / Avg: 200.67 / Max: 206.89 Min: 202.37 / Avg: 204.69 / Max: 206.02 Min: 200.25 / Avg: 202.15 / Max: 204.16 1. (CC) gcc options: -O2
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 79.79 79.09 78.83 79.79 80.37 80.29 78.90 80.18 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations GRAPHITE LTO LTO GRAPHITE 15 30 45 60 75 Min: 79.75 / Avg: 79.79 / Max: 79.82 Min: 78.99 / Avg: 79.09 / Max: 79.23 Min: 78.77 / Avg: 78.83 / Max: 78.89 Min: 79.74 / Avg: 79.79 / Max: 79.83 Min: 80.18 / Avg: 80.37 / Max: 80.55 Min: 80.18 / Avg: 80.29 / Max: 80.44 Min: 78.82 / Avg: 78.9 / Max: 79.03 Min: 80.07 / Avg: 80.18 / Max: 80.35 1. (CC) gcc options: -march=native -pipe -O3 -lm
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: http GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 18.36, N = 3 SE +/- 25.13, N = 3 SE +/- 13.05, N = 3 SE +/- 5.61, N = 3 SE +/- 1.45, N = 3 SE +/- 19.46, N = 3 SE +/- 29.02, N = 3 SE +/- 13.61, N = 3 3583 3566 3608 3544 3573 3539 3561 3580 Result Confidence
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: http GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3554 / Avg: 3583 / Max: 3617 Min: 3523 / Avg: 3565.67 / Max: 3610 Min: 3582 / Avg: 3608 / Max: 3623 Min: 3536 / Avg: 3544.33 / Max: 3555 Min: 3571 / Avg: 3573.33 / Max: 3576 Min: 3508 / Avg: 3539.33 / Max: 3575 Min: 3523 / Avg: 3561 / Max: 3618 Min: 3554 / Avg: 3580 / Max: 3600
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Render Test GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 SE +/- 0.12, N = 2 SE +/- 0.37, N = 2 SE +/- 0.11, N = 2 SE +/- 0.11, N = 2 SE +/- 0.29, N = 2 SE +/- 0.06, N = 2 SE +/- 0.03, N = 2 SE +/- 0.01, N = 2 33.30 33.26 33.49 33.06 32.85 32.85 33.06 33.12 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Render Test GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 Min: 33.18 / Avg: 33.3 / Max: 33.42 Min: 32.89 / Avg: 33.26 / Max: 33.62 Min: 33.38 / Avg: 33.49 / Max: 33.6 Min: 32.95 / Avg: 33.06 / Max: 33.17 Min: 32.56 / Avg: 32.85 / Max: 33.13 Min: 32.78 / Avg: 32.85 / Max: 32.91 Min: 33.04 / Avg: 33.06 / Max: 33.09 Min: 33.12 / Avg: 33.12 / Max: 33.13
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Set GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K SE +/- 382.34, N = 3 SE +/- 271.78, N = 3 SE +/- 202.88, N = 3 SE +/- 234.87, N = 3 SE +/- 267.06, N = 3 SE +/- 417.62, N = 3 SE +/- 501.40, N = 3 SE +/- 189.27, N = 3 76337 76973 76110 76123 75529 76206 76431 76448 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Set GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K Min: 75777.1 / Avg: 76337.2 / Max: 77068.1 Min: 76596.6 / Avg: 76973.27 / Max: 77501 Min: 75799.5 / Avg: 76109.63 / Max: 76491.3 Min: 75673 / Avg: 76123.2 / Max: 76464.4 Min: 75221.7 / Avg: 75529.23 / Max: 76061.2 Min: 75624.6 / Avg: 76205.8 / Max: 77015.9 Min: 75485.6 / Avg: 76430.7 / Max: 77193.6 Min: 76069.9 / Avg: 76448.43 / Max: 76638.9 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 16.92 17.11 17.00 17.24 17.07 17.13 17.07 17.16 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 16.86 / Avg: 16.92 / Max: 16.97 Min: 17.07 / Avg: 17.11 / Max: 17.16 Min: 16.9 / Avg: 17 / Max: 17.07 Min: 17.13 / Avg: 17.24 / Max: 17.39 Min: 17.02 / Avg: 17.07 / Max: 17.12 Min: 17.11 / Avg: 17.13 / Max: 17.16 Min: 17.03 / Avg: 17.07 / Max: 17.12 Min: 17.14 / Avg: 17.16 / Max: 17.2 1. (CC) gcc options: -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 58.24, N = 3 SE +/- 93.28, N = 3 SE +/- 27.58, N = 3 SE +/- 63.39, N = 3 SE +/- 38.13, N = 3 SE +/- 67.01, N = 3 SE +/- 61.86, N = 3 SE +/- 66.41, N = 3 16135 15933 16134 16099 16234 16142 16179 16180 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 16061.23 / Avg: 16134.87 / Max: 16249.84 Min: 15809.3 / Avg: 15933.04 / Max: 16115.83 Min: 16078.38 / Avg: 16133.53 / Max: 16161.62 Min: 15972.05 / Avg: 16098.77 / Max: 16165.7 Min: 16158.56 / Avg: 16233.88 / Max: 16281.88 Min: 16009.01 / Avg: 16142.13 / Max: 16222.04 Min: 16057.2 / Avg: 16179.45 / Max: 16257.07 Min: 16050.16 / Avg: 16179.87 / Max: 16269.46 1. (CC) gcc options: -O2
CoMD OpenCL CoMD benchmark in OpenCL via GPUOpen. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us/atom/task, More Is Better CoMD OpenCL 2017-07-06 Average Atom Update Rate GRAPHITE LTO LTO GRAPHITE 0.6143 1.2286 1.8429 2.4572 3.0715 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 2.69 2.69 2.72 2.73 2.69 2.71 2.68 2.68 1. (CC) gcc options: -std=c99 -O5 -lm -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org us/atom/task, More Is Better CoMD OpenCL 2017-07-06 Average Atom Update Rate GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.67 / Avg: 2.69 / Max: 2.72 Min: 2.67 / Avg: 2.69 / Max: 2.71 Min: 2.67 / Avg: 2.72 / Max: 2.79 Min: 2.68 / Avg: 2.73 / Max: 2.8 Min: 2.67 / Avg: 2.69 / Max: 2.71 Min: 2.68 / Avg: 2.71 / Max: 2.75 Min: 2.67 / Avg: 2.68 / Max: 2.68 Min: 2.67 / Avg: 2.68 / Max: 2.7 1. (CC) gcc options: -std=c99 -O5 -lm -lOpenCL
Gzip Compression This test measures the time needed to archive/compress two copies of the Linux 4.13 kernel source tree using Gzip compression. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Gzip Compression Linux Source Tree Archiving To .tar.gz GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 33.61 33.59 33.86 33.95 33.65 33.33 33.82 33.72 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Gzip Compression Linux Source Tree Archiving To .tar.gz GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 Min: 33.6 / Avg: 33.61 / Max: 33.64 Min: 33.58 / Avg: 33.59 / Max: 33.6 Min: 33.82 / Avg: 33.86 / Max: 33.9 Min: 33.92 / Avg: 33.95 / Max: 33.99 Min: 33.63 / Avg: 33.65 / Max: 33.67 Min: 33.32 / Avg: 33.33 / Max: 33.34 Min: 33.8 / Avg: 33.82 / Max: 33.84 Min: 33.69 / Avg: 33.72 / Max: 33.77
CloverLeaf CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm8192.in input file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics GRAPHITE LTO LTO GRAPHITE 0.3713 0.7426 1.1139 1.4852 1.8565 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 1.64 1.64 1.63 1.65 1.64 1.62 1.63 1.63 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.62 / Avg: 1.64 / Max: 1.65 Min: 1.62 / Avg: 1.64 / Max: 1.65 Min: 1.62 / Avg: 1.63 / Max: 1.65 Min: 1.63 / Avg: 1.65 / Max: 1.66 Min: 1.62 / Avg: 1.64 / Max: 1.66 Min: 1.61 / Avg: 1.62 / Max: 1.63 Min: 1.62 / Avg: 1.63 / Max: 1.66 Min: 1.62 / Avg: 1.63 / Max: 1.64 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Get GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 895.94, N = 3 SE +/- 756.08, N = 3 SE +/- 328.44, N = 3 SE +/- 233.79, N = 3 SE +/- 176.25, N = 3 SE +/- 696.74, N = 3 SE +/- 1243.20, N = 3 SE +/- 710.69, N = 3 111960 111798 112940 113860 113528 112397 113824 113117 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Get GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 110209.4 / Avg: 111959.83 / Max: 113166.9 Min: 110392.1 / Avg: 111798.43 / Max: 112982.9 Min: 112522.2 / Avg: 112940.03 / Max: 113587.9 Min: 113414.9 / Avg: 113860.27 / Max: 114206.3 Min: 113291.1 / Avg: 113527.57 / Max: 113872.2 Min: 111008.9 / Avg: 112397.2 / Max: 113195.3 Min: 112220.3 / Avg: 113823.77 / Max: 116271.2 Min: 111712.2 / Avg: 113116.53 / Max: 114008.8 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
Cython benchmark Stress benchmark tests to measure time consumed by cython code Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 22.59 22.71 22.55 22.60 22.33 22.30 22.33 22.65 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 Min: 22.57 / Avg: 22.59 / Max: 22.6 Min: 22.6 / Avg: 22.71 / Max: 22.87 Min: 22.48 / Avg: 22.55 / Max: 22.64 Min: 22.48 / Avg: 22.6 / Max: 22.67 Min: 22.28 / Avg: 22.33 / Max: 22.41 Min: 22.24 / Avg: 22.3 / Max: 22.41 Min: 22.32 / Avg: 22.33 / Max: 22.34 Min: 22.59 / Avg: 22.65 / Max: 22.76
eSpeak Speech Engine This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better eSpeak Speech Engine 1.48.04 Text-To-Speech Synthesis GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 SE +/- 0.22, N = 3 SE +/- 0.23, N = 3 SE +/- 0.03, N = 3 SE +/- 0.25, N = 3 SE +/- 0.20, N = 3 SE +/- 0.26, N = 3 SE +/- 0.20, N = 3 SE +/- 0.22, N = 3 25.61 25.75 25.35 25.68 25.69 25.81 25.70 25.68 1. (CXX) g++ options: -lstdc++ -lespeak Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better eSpeak Speech Engine 1.48.04 Text-To-Speech Synthesis GRAPHITE LTO LTO GRAPHITE 6 12 18 24 30 Min: 25.18 / Avg: 25.61 / Max: 25.86 Min: 25.29 / Avg: 25.75 / Max: 26.02 Min: 25.3 / Avg: 25.35 / Max: 25.41 Min: 25.18 / Avg: 25.68 / Max: 25.95 Min: 25.31 / Avg: 25.69 / Max: 25.97 Min: 25.3 / Avg: 25.81 / Max: 26.13 Min: 25.32 / Avg: 25.7 / Max: 25.95 Min: 25.24 / Avg: 25.68 / Max: 25.92 1. (CXX) g++ options: -lstdc++ -lespeak
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex GRAPHITE LTO LTO GRAPHITE 0.8843 1.7686 2.6529 3.5372 4.4215 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 3.93 3.91 3.90 3.86 3.90 3.93 3.91 3.87 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.92 / Avg: 3.93 / Max: 3.93 Min: 3.91 / Avg: 3.91 / Max: 3.92 Min: 3.89 / Avg: 3.9 / Max: 3.9 Min: 3.86 / Avg: 3.86 / Max: 3.88 Min: 3.89 / Avg: 3.9 / Max: 3.91 Min: 3.92 / Avg: 3.93 / Max: 3.94 Min: 3.89 / Avg: 3.91 / Max: 3.92 Min: 3.86 / Avg: 3.87 / Max: 3.87 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lglut -lGL -lGLU
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Math Library GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 SE +/- 0.20, N = 3 SE +/- 0.05, N = 3 SE +/- 0.61, N = 3 SE +/- 2.60, N = 3 SE +/- 0.50, N = 3 SE +/- 1.19, N = 3 SE +/- 0.18, N = 3 SE +/- 0.43, N = 3 278 278 279 283 278 280 278 279 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Math Library GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 277.31 / Avg: 277.62 / Max: 277.99 Min: 277.56 / Avg: 277.62 / Max: 277.71 Min: 277.61 / Avg: 278.68 / Max: 279.73 Min: 278.55 / Avg: 282.97 / Max: 287.55 Min: 277.47 / Avg: 278 / Max: 278.99 Min: 277.84 / Avg: 279.76 / Max: 281.94 Min: 277.97 / Avg: 278.32 / Max: 278.57 Min: 278.69 / Avg: 279.21 / Max: 280.07 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 6.89 6.88 6.99 6.97 6.96 6.87 6.91 6.97 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 6.83 / Avg: 6.89 / Max: 7 Min: 6.83 / Avg: 6.88 / Max: 6.95 Min: 6.97 / Avg: 6.99 / Max: 7.01 Min: 6.88 / Avg: 6.97 / Max: 7.07 Min: 6.92 / Avg: 6.96 / Max: 7.03 Min: 6.78 / Avg: 6.87 / Max: 6.98 Min: 6.78 / Avg: 6.91 / Max: 7 Min: 6.91 / Avg: 6.97 / Max: 7.03 1. (CC) gcc options: -lpthread -march=native -pipe -O3
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Composite GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 1.95, N = 3 SE +/- 1.06, N = 3 SE +/- 1.79, N = 3 SE +/- 1.01, N = 3 SE +/- 1.31, N = 3 SE +/- 0.86, N = 3 SE +/- 1.18, N = 3 SE +/- 1.78, N = 3 1574 1576 1592 1568 1575 1595 1589 1573 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Composite GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 Min: 1570.15 / Avg: 1573.72 / Max: 1576.85 Min: 1574.34 / Avg: 1575.75 / Max: 1577.82 Min: 1588.2 / Avg: 1591.78 / Max: 1593.65 Min: 1566.27 / Avg: 1568.28 / Max: 1569.31 Min: 1573.07 / Avg: 1574.64 / Max: 1577.24 Min: 1593.45 / Avg: 1594.93 / Max: 1596.42 Min: 1586.39 / Avg: 1588.67 / Max: 1590.35 Min: 1569.67 / Avg: 1572.96 / Max: 1575.77 1. (CC) gcc options: -lm -ldl
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 SE +/- 2.03, N = 3 SE +/- 2.90, N = 3 SE +/- 4.20, N = 3 SE +/- 3.67, N = 3 SE +/- 5.77, N = 3 SE +/- 4.13, N = 3 SE +/- 3.33, N = 3 SE +/- 3.91, N = 3 354 354 353 353 358 359 357 358 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -fPIC -O2 -flto -fvisibility=hidden -pie -rdynamic -lpthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 Min: 350.06 / Avg: 353.94 / Max: 356.93 Min: 348.03 / Avg: 353.61 / Max: 357.78 Min: 345.03 / Avg: 353.39 / Max: 358.21 Min: 346.22 / Avg: 353.23 / Max: 358.64 Min: 347.42 / Avg: 358.39 / Max: 366.97 Min: 350.26 / Avg: 358.52 / Max: 362.76 Min: 350.26 / Avg: 356.78 / Max: 361.23 Min: 350.26 / Avg: 358.08 / Max: 362.32 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -fPIC -O2 -flto -fvisibility=hidden -pie -rdynamic -lpthread -lrt
m-queens A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 37.71 37.70 37.10 37.13 37.72 37.72 37.09 37.12 1. (CXX) g++ options: -fopenmp -march=native -pipe -O3 -O2 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 Min: 37.68 / Avg: 37.71 / Max: 37.73 Min: 37.69 / Avg: 37.7 / Max: 37.7 Min: 37.07 / Avg: 37.1 / Max: 37.15 Min: 37.06 / Avg: 37.13 / Max: 37.17 Min: 37.71 / Avg: 37.72 / Max: 37.73 Min: 37.68 / Avg: 37.72 / Max: 37.74 Min: 37.06 / Avg: 37.09 / Max: 37.13 Min: 37.09 / Avg: 37.12 / Max: 37.13 1. (CXX) g++ options: -fopenmp -march=native -pipe -O3 -O2
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 40K 80K 120K 160K 200K SE +/- 886.81, N = 3 SE +/- 615.35, N = 3 SE +/- 1230.69, N = 3 SE +/- 2076.25, N = 3 SE +/- 2131.63, N = 3 SE +/- 2048.00, N = 3 SE +/- 170.67, N = 3 SE +/- 2407.55, N = 3 173824 173483 170923 171435 172800 172288 171947 172117 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K Min: 172288 / Avg: 173824 / Max: 175360 Min: 172288 / Avg: 173482.67 / Max: 174336 Min: 169216 / Avg: 170922.67 / Max: 173312 Min: 167680 / Avg: 171434.67 / Max: 174848 Min: 168704 / Avg: 172800 / Max: 175872 Min: 170240 / Avg: 172288 / Max: 176384 Min: 171776 / Avg: 171946.67 / Max: 172288 Min: 169216 / Avg: 172117.33 / Max: 176896 1. (CC) gcc options: -O2 -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 223.47, N = 3 SE +/- 88.63, N = 3 SE +/- 146.36, N = 3 SE +/- 206.08, N = 3 SE +/- 194.64, N = 3 SE +/- 141.63, N = 3 SE +/- 130.87, N = 3 SE +/- 38.69, N = 3 34736 34603 34409 34634 34589 34558 34191 34770 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 34376.26 / Avg: 34735.93 / Max: 35145.52 Min: 34443.32 / Avg: 34602.84 / Max: 34749.56 Min: 34161.8 / Avg: 34409.08 / Max: 34668.38 Min: 34226.89 / Avg: 34634.11 / Max: 34892.83 Min: 34200.6 / Avg: 34589.4 / Max: 34800.34 Min: 34277.3 / Avg: 34557.62 / Max: 34733.06 Min: 33936.5 / Avg: 34191.32 / Max: 34370.49 Min: 34717.75 / Avg: 34769.7 / Max: 34845.34 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 36.07, N = 3 SE +/- 37.46, N = 3 SE +/- 48.84, N = 3 SE +/- 65.47, N = 3 SE +/- 89.99, N = 3 SE +/- 26.56, N = 3 SE +/- 20.68, N = 3 SE +/- 64.53, N = 3 32575 32243 32296 32669 32716 32787 32748 32765 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 32506.39 / Avg: 32575.41 / Max: 32628.09 Min: 32203.79 / Avg: 32243.25 / Max: 32318.13 Min: 32198.73 / Avg: 32296.37 / Max: 32347.74 Min: 32592.78 / Avg: 32668.6 / Max: 32798.96 Min: 32607.31 / Avg: 32716.32 / Max: 32894.84 Min: 32735 / Avg: 32786.93 / Max: 32822.62 Min: 32708.86 / Avg: 32748.12 / Max: 32779.01 Min: 32643.69 / Avg: 32765.11 / Max: 32863.7 1. (CC) gcc options: -O2
FS-Mark FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 3.24, N = 3 SE +/- 1.11, N = 3 SE +/- 2.07, N = 3 SE +/- 1.97, N = 3 SE +/- 1.81, N = 3 SE +/- 1.05, N = 3 SE +/- 0.78, N = 3 SE +/- 0.89, N = 3 2556 2574 2582 2568 2559 2558 2540 2566 1. (CC) gcc options: -static Result Confidence
O2 O3 OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 2550.3 / Avg: 2555.77 / Max: 2561.5 Min: 2572.4 / Avg: 2574.43 / Max: 2576.2 Min: 2578.9 / Avg: 2581.63 / Max: 2585.7 Min: 2564.4 / Avg: 2567.7 / Max: 2571.2 Min: 2556 / Avg: 2558.83 / Max: 2562.2 Min: 2556.6 / Avg: 2557.8 / Max: 2559.9 Min: 2538.5 / Avg: 2540.07 / Max: 2540.9 Min: 2564.6 / Avg: 2566.3 / Max: 2567.6 1. (CC) gcc options: -static
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 1.00, N = 3 SE +/- 1.25, N = 15 SE +/- 1.53, N = 3 SE +/- 1.53, N = 15 SE +/- 1.48, N = 15 SE +/- 1.33, N = 3 SE +/- 2.40, N = 3 122 121 123 121 122 123 123 122 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 121 / Avg: 122 / Max: 124 Min: 116 / Avg: 121.4 / Max: 131 Min: 118 / Avg: 121 / Max: 123 Min: 117 / Avg: 121.87 / Max: 137 Min: 116 / Avg: 122.8 / Max: 137 Min: 120 / Avg: 122.67 / Max: 124 Min: 117 / Avg: 121.67 / Max: 125 1. (CC) gcc options: -O2 -lpthread
Apache Benchmark This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K SE +/- 44.34, N = 3 SE +/- 46.17, N = 3 SE +/- 74.05, N = 3 SE +/- 53.26, N = 3 SE +/- 35.61, N = 3 SE +/- 76.05, N = 3 SE +/- 91.18, N = 3 SE +/- 149.61, N = 3 38423 38140 38201 38628 38386 38189 38541 38755 1. (CC) gcc options: -shared -fPIC -pthread -march=native -O3 Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 38351.54 / Avg: 38423.45 / Max: 38504.36 Min: 38074.96 / Avg: 38139.62 / Max: 38229.04 Min: 38121 / Avg: 38201.12 / Max: 38349.05 Min: 38529.34 / Avg: 38627.65 / Max: 38712.33 Min: 38319.75 / Avg: 38385.84 / Max: 38441.87 Min: 38040.17 / Avg: 38189.2 / Max: 38290.02 Min: 38359.78 / Avg: 38541.22 / Max: 38647.79 Min: 38503.81 / Avg: 38755.15 / Max: 39021.43 1. (CC) gcc options: -shared -fPIC -pthread -march=native -O3
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create GRAPHITE LTO LTO GRAPHITE 120 240 360 480 600 SE +/- 2.51, N = 3 SE +/- 3.41, N = 3 SE +/- 2.39, N = 3 SE +/- 3.71, N = 3 SE +/- 1.71, N = 3 SE +/- 2.15, N = 3 SE +/- 2.37, N = 3 SE +/- 2.96, N = 3 565 561 568 559 561 559 564 564 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create GRAPHITE LTO LTO GRAPHITE 100 200 300 400 500 Min: 560.21 / Avg: 564.96 / Max: 568.76 Min: 556 / Avg: 560.71 / Max: 567.34 Min: 563.06 / Avg: 567.84 / Max: 570.27 Min: 551.87 / Avg: 558.88 / Max: 564.49 Min: 558.79 / Avg: 561.16 / Max: 564.49 Min: 554.58 / Avg: 558.81 / Max: 561.64 Min: 558.79 / Avg: 563.54 / Max: 565.91 Min: 558.79 / Avg: 564.49 / Max: 568.76
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 1100 2200 3300 4400 5500 SE +/- 6.19, N = 4 SE +/- 15.54, N = 4 SE +/- 21.41, N = 4 SE +/- 23.10, N = 4 SE +/- 7.74, N = 4 SE +/- 30.02, N = 4 SE +/- 27.01, N = 4 SE +/- 23.69, N = 4 5034 5080 5090 5065 5010 5047 5064 5021 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Dense LU Matrix Factorization GRAPHITE LTO LTO GRAPHITE 900 1800 2700 3600 4500 Min: 5015 / Avg: 5033.51 / Max: 5040.46 Min: 5035.81 / Avg: 5080.45 / Max: 5101.67 Min: 5047.44 / Avg: 5090.05 / Max: 5130.42 Min: 5021.92 / Avg: 5065.31 / Max: 5106.44 Min: 4992.08 / Avg: 5010.43 / Max: 5028.85 Min: 5001.22 / Avg: 5047.39 / Max: 5135.25 Min: 4985.24 / Avg: 5063.67 / Max: 5108.82 Min: 4976.16 / Avg: 5021.1 / Max: 5085.04
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K SE +/- 357.29, N = 3 SE +/- 400.03, N = 3 SE +/- 859.84, N = 3 SE +/- 90.01, N = 3 SE +/- 369.84, N = 3 SE +/- 370.02, N = 3 SE +/- 184.32, N = 3 SE +/- 491.80, N = 3 60598 61101 60646 60408 60580 61107 60833 61336 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K Min: 60039 / Avg: 60598 / Max: 61263 Min: 60429 / Avg: 61101 / Max: 61813 Min: 59041 / Avg: 60646.33 / Max: 61983 Min: 60228 / Avg: 60408 / Max: 60500 Min: 59880 / Avg: 60580 / Max: 61137 Min: 60643 / Avg: 61106.67 / Max: 61838 Min: 60587 / Avg: 60833.33 / Max: 61194 Min: 60385 / Avg: 61336 / Max: 62029 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Eclipse GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 59.66, N = 4 SE +/- 53.73, N = 4 SE +/- 36.12, N = 4 SE +/- 40.31, N = 4 SE +/- 26.40, N = 4 SE +/- 17.58, N = 4 SE +/- 52.12, N = 4 SE +/- 59.57, N = 4 16510 16434 16416 16556 16501 16588 16585 16667 Result Confidence
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Eclipse GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 16408 / Avg: 16509.5 / Max: 16658 Min: 16319 / Avg: 16433.75 / Max: 16541 Min: 16318 / Avg: 16416.25 / Max: 16473 Min: 16443 / Avg: 16556 / Max: 16621 Min: 16450 / Avg: 16500.75 / Max: 16573 Min: 16545 / Avg: 16587.75 / Max: 16631 Min: 16461 / Avg: 16585 / Max: 16697 Min: 16556 / Avg: 16667.25 / Max: 16833
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: SP.A GRAPHITE LTO LTO GRAPHITE 900 1800 2700 3600 4500 SE +/- 38.26, N = 3 SE +/- 20.07, N = 3 SE +/- 40.41, N = 3 SE +/- 8.41, N = 3 SE +/- 46.73, N = 3 SE +/- 22.51, N = 3 SE +/- 24.81, N = 3 SE +/- 22.50, N = 3 3937 3964 3977 3939 3990 3966 3935 3930 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: SP.A GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 3897.48 / Avg: 3936.6 / Max: 4013.11 Min: 3930.82 / Avg: 3963.68 / Max: 4000.08 Min: 3896.42 / Avg: 3976.72 / Max: 4024.84 Min: 3927.6 / Avg: 3938.81 / Max: 3955.27 Min: 3903.64 / Avg: 3990.26 / Max: 4063.96 Min: 3928.19 / Avg: 3966.01 / Max: 4006.06 Min: 3899.19 / Avg: 3934.95 / Max: 3982.63 Min: 3898.52 / Avg: 3930.13 / Max: 3973.66 1. 3.3
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.5 1080p 8-bit YUV To AV1 Video Encode GRAPHITE LTO LTO GRAPHITE 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.14, N = 3 SE +/- 0.26, N = 3 SE +/- 0.19, N = 3 SE +/- 0.20, N = 3 SE +/- 0.19, N = 3 SE +/- 0.11, N = 3 SE +/- 0.37, N = 3 55.19 54.62 54.91 55.01 54.63 55.28 55.44 55.03 1. (CXX) g++ options: -march=native -pipe -O3 -pie -lpthread -lm Result Confidence
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.5 1080p 8-bit YUV To AV1 Video Encode GRAPHITE LTO LTO GRAPHITE 11 22 33 44 55 Min: 54.96 / Avg: 55.19 / Max: 55.42 Min: 54.42 / Avg: 54.62 / Max: 54.89 Min: 54.39 / Avg: 54.91 / Max: 55.22 Min: 54.63 / Avg: 55.01 / Max: 55.25 Min: 54.26 / Avg: 54.63 / Max: 54.95 Min: 54.9 / Avg: 55.28 / Max: 55.51 Min: 55.22 / Avg: 55.44 / Max: 55.56 Min: 54.39 / Avg: 55.03 / Max: 55.66 1. (CXX) g++ options: -march=native -pipe -O3 -pie -lpthread -lm
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: tanh GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.14, N = 7 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 10.70 10.69 10.69 10.70 10.84 10.68 10.69 10.68 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: tanh GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10.69 / Avg: 10.7 / Max: 10.7 Min: 10.68 / Avg: 10.69 / Max: 10.69 Min: 10.68 / Avg: 10.69 / Max: 10.69 Min: 10.7 / Avg: 10.7 / Max: 10.7 Min: 10.69 / Avg: 10.84 / Max: 11.68 Min: 10.67 / Avg: 10.68 / Max: 10.69 Min: 10.68 / Avg: 10.69 / Max: 10.69 Min: 10.66 / Avg: 10.68 / Max: 10.69
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K SE +/- 371.96, N = 3 SE +/- 443.41, N = 3 SE +/- 1048.59, N = 3 SE +/- 644.25, N = 3 SE +/- 519.06, N = 3 SE +/- 1065.81, N = 3 SE +/- 341.33, N = 3 SE +/- 814.03, N = 3 75477 75648 75733 75392 74965 74624 75563 75733 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 24 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K Min: 74880 / Avg: 75477.33 / Max: 76160 Min: 74880 / Avg: 75648 / Max: 76416 Min: 74112 / Avg: 75733.33 / Max: 77696 Min: 74112 / Avg: 75392 / Max: 76160 Min: 74112 / Avg: 74965.33 / Max: 75904 Min: 73088 / Avg: 74624 / Max: 76672 Min: 74880 / Avg: 75562.67 / Max: 75904 Min: 74112 / Avg: 75733.33 / Max: 76672 1. (CC) gcc options: -O2 -lpthread
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Prepend GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K SE +/- 730.29, N = 3 SE +/- 737.81, N = 3 SE +/- 477.37, N = 3 SE +/- 604.44, N = 3 SE +/- 671.62, N = 3 SE +/- 283.44, N = 3 SE +/- 97.01, N = 3 SE +/- 417.94, N = 3 78449 78873 78372 79528 79023 78549 79349 78756 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Prepend GRAPHITE LTO LTO GRAPHITE 14K 28K 42K 56K 70K Min: 77263.2 / Avg: 78448.97 / Max: 79780.4 Min: 77870.9 / Avg: 78872.57 / Max: 80311.8 Min: 77661.2 / Avg: 78371.77 / Max: 79279.3 Min: 78321 / Avg: 79527.67 / Max: 80194.2 Min: 77830.7 / Avg: 79022.97 / Max: 80154.9 Min: 78159.7 / Avg: 78549.13 / Max: 79100.6 Min: 79200.3 / Avg: 79348.8 / Max: 79531.2 Min: 78215 / Avg: 78755.83 / Max: 79578.2 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
Result
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Add GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K SE +/- 187.94, N = 3 SE +/- 347.01, N = 3 SE +/- 198.78, N = 3 SE +/- 145.84, N = 3 SE +/- 28.57, N = 3 SE +/- 339.42, N = 3 SE +/- 270.14, N = 3 SE +/- 809.20, N = 3 76812 76179 75844 76835 75844 75929 76962 76915 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Add GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K Min: 76441.1 / Avg: 76812.23 / Max: 77049.4 Min: 75517.4 / Avg: 76179.03 / Max: 76691.3 Min: 75473.3 / Avg: 75843.67 / Max: 76154 Min: 76561.1 / Avg: 76834.63 / Max: 77059.1 Min: 75786.8 / Avg: 75843.8 / Max: 75875.8 Min: 75254.4 / Avg: 75929.1 / Max: 76331.2 Min: 76482.7 / Avg: 76962.03 / Max: 77417.6 Min: 75393.8 / Avg: 76915.23 / Max: 78153.8 1. (CC) gcc options: -march=native -pipe -O3 -lm -rdynamic
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org us, Fewer Is Better clpeak OpenCL Test: Kernel Latency GRAPHITE LTO LTO GRAPHITE 1.0845 2.169 3.2535 4.338 5.4225 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 4.79 4.82 4.76 4.77 4.76 4.76 4.76 4.75 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org us, Fewer Is Better clpeak OpenCL Test: Kernel Latency GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 4.78 / Avg: 4.79 / Max: 4.79 Min: 4.81 / Avg: 4.82 / Max: 4.82 Min: 4.76 / Avg: 4.76 / Max: 4.77 Min: 4.75 / Avg: 4.77 / Max: 4.79 Min: 4.74 / Avg: 4.76 / Max: 4.78 Min: 4.74 / Avg: 4.76 / Max: 4.77 Min: 4.73 / Avg: 4.76 / Max: 4.8 Min: 4.75 / Avg: 4.75 / Max: 4.75 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Process GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.07, N = 3 SE +/- 0.18, N = 3 SE +/- 0.06, N = 3 SE +/- 0.13, N = 3 SE +/- 0.18, N = 3 SE +/- 0.04, N = 3 22.73 22.86 22.84 22.74 22.66 22.74 22.78 22.53 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Process GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 Min: 22.69 / Avg: 22.73 / Max: 22.78 Min: 22.58 / Avg: 22.86 / Max: 23.1 Min: 22.77 / Avg: 22.84 / Max: 22.98 Min: 22.39 / Avg: 22.74 / Max: 22.99 Min: 22.55 / Avg: 22.66 / Max: 22.78 Min: 22.51 / Avg: 22.74 / Max: 22.96 Min: 22.44 / Avg: 22.78 / Max: 23.07 Min: 22.46 / Avg: 22.53 / Max: 22.59 1. (CC) gcc options: -lpthread -march=native -pipe -O3
7-Zip Compression This is a test of 7-Zip using p7zip with its integrated benchmark feature or upstream 7-Zip for the Windows x64 build. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 326.76, N = 3 SE +/- 269.21, N = 3 SE +/- 366.05, N = 3 SE +/- 500.30, N = 3 SE +/- 328.13, N = 3 SE +/- 108.27, N = 3 SE +/- 385.01, N = 3 SE +/- 250.71, N = 3 119324 119307 119702 118911 118713 118640 118132 117979 1. (CXX) g++ options: -pipe -lpthread Result Confidence
O2 O3 OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 118975 / Avg: 119324 / Max: 119977 Min: 118780 / Avg: 119307 / Max: 119666 Min: 119326 / Avg: 119702 / Max: 120434 Min: 118265 / Avg: 118911.33 / Max: 119896 Min: 118067 / Avg: 118713.33 / Max: 119135 Min: 118450 / Avg: 118639.67 / Max: 118825 Min: 117478 / Avg: 118132 / Max: 118811 Min: 117509 / Avg: 117979.33 / Max: 118365 1. (CXX) g++ options: -pipe -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 131.59, N = 3 SE +/- 156.65, N = 3 SE +/- 350.13, N = 3 SE +/- 48.94, N = 3 SE +/- 311.21, N = 3 SE +/- 184.15, N = 3 SE +/- 72.28, N = 3 SE +/- 253.18, N = 3 31331 30903 30905 31018 31150 31353 31164 31311 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K Min: 31083.05 / Avg: 31331.42 / Max: 31530.98 Min: 30592.73 / Avg: 30903.3 / Max: 31094.38 Min: 30206.49 / Avg: 30904.75 / Max: 31299.67 Min: 30936.56 / Avg: 31017.93 / Max: 31105.71 Min: 30688.08 / Avg: 31149.73 / Max: 31742.1 Min: 31098.15 / Avg: 31352.93 / Max: 31710.64 Min: 31037.83 / Avg: 31164.11 / Max: 31288.19 Min: 30847.09 / Avg: 31311.43 / Max: 31718.5 1. (CC) gcc options: -O2
BlogBench BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 258.32, N = 3 SE +/- 289.89, N = 3 SE +/- 152.69, N = 3 SE +/- 601.54, N = 3 SE +/- 144.80, N = 3 SE +/- 217.73, N = 3 SE +/- 745.12, N = 3 SE +/- 91.62, N = 3 41331 41410 41045 40873 41466 41093 41094 40914 1. (CC) gcc options: -march=native -pipe -O3 -pthread Result Confidence
O2 O3 OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 40822 / Avg: 41331.33 / Max: 41661 Min: 40832 / Avg: 41410.33 / Max: 41735 Min: 40809 / Avg: 41045.33 / Max: 41331 Min: 40012 / Avg: 40872.67 / Max: 42031 Min: 41289 / Avg: 41466 / Max: 41753 Min: 40858 / Avg: 41093 / Max: 41528 Min: 40167 / Avg: 41094 / Max: 42568 Min: 40779 / Avg: 40914.33 / Max: 41089 1. (CC) gcc options: -march=native -pipe -O3 -pthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 15K 30K 45K 60K 75K SE +/- 228.80, N = 3 SE +/- 160.16, N = 3 SE +/- 353.68, N = 3 SE +/- 293.38, N = 3 SE +/- 278.46, N = 3 SE +/- 375.31, N = 3 SE +/- 85.93, N = 3 SE +/- 128.80, N = 3 68738 69081 68789 68985 69248 69063 68659 68265 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K Min: 68457.21 / Avg: 68738.27 / Max: 69191.53 Min: 68814.89 / Avg: 69081.02 / Max: 69368.47 Min: 68102.09 / Avg: 68789.27 / Max: 69278.13 Min: 68412.61 / Avg: 68984.67 / Max: 69383.75 Min: 68841.49 / Avg: 69247.96 / Max: 69780.91 Min: 68385.2 / Avg: 69063.01 / Max: 69681.2 Min: 68555.74 / Avg: 68659.38 / Max: 68829.93 Min: 68136.07 / Avg: 68264.88 / Max: 68522.48 1. (CC) gcc options: -O2
Apache Siege This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 1 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 128.63, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 114.56, N = 3 SE +/- 0.00, N = 3 SE +/- 128.63, N = 3 SE +/- 65.21, N = 3 SE +/- 0.00, N = 3 13956 13889 14085 14086 13889 13956 14019 13889 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto Result Confidence
O2 O3 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 1 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 13698.63 / Avg: 13955.88 / Max: 14084.51 Min: 13888.89 / Avg: 13888.89 / Max: 13888.89 Min: 14084.51 / Avg: 14084.51 / Max: 14084.51 Min: 13888.89 / Avg: 14086.37 / Max: 14285.71 Min: 13888.89 / Avg: 13888.89 / Max: 13888.89 Min: 13698.63 / Avg: 13955.88 / Max: 14084.51 Min: 13888.89 / Avg: 14019.3 / Max: 14084.51 Min: 13888.89 / Avg: 13888.89 / Max: 13888.89 1. (CC) gcc options: -march=native -pipe -O3 -lpthread -ldl -lssl -lcrypto
x265 This is a simple test of the x265 encoder run on the CPU with a sample 1080p video file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 SE +/- 0.07, N = 3 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.23, N = 3 SE +/- 0.04, N = 3 56.43 56.10 56.87 56.43 56.15 56.34 56.57 56.46 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lpthread -lrt -ldl -lnuma Result Confidence
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding GRAPHITE LTO LTO GRAPHITE 11 22 33 44 55 Min: 56.32 / Avg: 56.43 / Max: 56.55 Min: 55.95 / Avg: 56.1 / Max: 56.39 Min: 56.61 / Avg: 56.87 / Max: 57.05 Min: 56.4 / Avg: 56.43 / Max: 56.46 Min: 56.05 / Avg: 56.15 / Max: 56.3 Min: 56.23 / Avg: 56.34 / Max: 56.54 Min: 56.27 / Avg: 56.57 / Max: 57.03 Min: 56.4 / Avg: 56.46 / Max: 56.53 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 13.35, N = 3 SE +/- 60.46, N = 3 SE +/- 95.66, N = 3 SE +/- 160.11, N = 3 SE +/- 74.75, N = 3 SE +/- 67.45, N = 3 SE +/- 47.99, N = 3 SE +/- 63.88, N = 3 17247 17161 17142 17318 17102 17110 17086 17283 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 17225.72 / Avg: 17247.01 / Max: 17271.62 Min: 17043.37 / Avg: 17160.62 / Max: 17244.86 Min: 16970.22 / Avg: 17142.11 / Max: 17300.8 Min: 17129.19 / Avg: 17318.27 / Max: 17636.62 Min: 16998.67 / Avg: 17101.52 / Max: 17246.9 Min: 17010.81 / Avg: 17109.97 / Max: 17238.77 Min: 17000.93 / Avg: 17085.64 / Max: 17167.09 Min: 17158.75 / Avg: 17282.98 / Max: 17370.95 1. (CC) gcc options: -O2
Node.js Octane Benchmark A Node.js version of the JavaScript Octane Benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Score, More Is Better Node.js Octane Benchmark GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 55.11, N = 3 SE +/- 119.65, N = 3 SE +/- 5.78, N = 3 SE +/- 214.29, N = 3 SE +/- 74.70, N = 3 SE +/- 51.85, N = 3 SE +/- 131.34, N = 3 SE +/- 100.73, N = 3 50696 50954 51009 51080 51363 51346 51217 51345 1. Nodejs Result Confidence
O2 O3 OpenBenchmarking.org Score, More Is Better Node.js Octane Benchmark GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 50626 / Avg: 50696.33 / Max: 50805 Min: 50715 / Avg: 50953.67 / Max: 51088 Min: 50999 / Avg: 51009.33 / Max: 51019 Min: 50693 / Avg: 51080 / Max: 51433 Min: 51219 / Avg: 51362.67 / Max: 51470 Min: 51292 / Avg: 51346.33 / Max: 51450 Min: 50954 / Avg: 51216.67 / Max: 51350 Min: 51144 / Avg: 51345.33 / Max: 51452 1. Nodejs
FLAC Audio Encoding This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 9.25 9.16 9.17 9.17 9.20 9.18 9.15 9.13 1. (CXX) g++ options: -march=native -pipe -O3 -fvisibility=hidden -logg -lm Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 9.24 / Avg: 9.25 / Max: 9.26 Min: 9.15 / Avg: 9.16 / Max: 9.17 Min: 9.16 / Avg: 9.17 / Max: 9.18 Min: 9.16 / Avg: 9.17 / Max: 9.19 Min: 9.17 / Avg: 9.2 / Max: 9.25 Min: 9.16 / Avg: 9.18 / Max: 9.2 Min: 9.13 / Avg: 9.15 / Max: 9.17 Min: 9.09 / Avg: 9.13 / Max: 9.15 1. (CXX) g++ options: -march=native -pipe -O3 -fvisibility=hidden -logg -lm
Hierarchical INTegration This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: DOUBLE GRAPHITE LTO LTO GRAPHITE 200M 400M 600M 800M 1000M SE +/- 1129622.34, N = 3 SE +/- 296143.00, N = 3 SE +/- 3024118.72, N = 3 SE +/- 1620971.61, N = 3 SE +/- 3203089.79, N = 3 SE +/- 1939878.89, N = 3 SE +/- 1986693.57, N = 3 SE +/- 949353.12, N = 3 955777557 953591098 954203727 956651722 956415576 958824024 953587375 966044516 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: DOUBLE GRAPHITE LTO LTO GRAPHITE 200M 400M 600M 800M 1000M Min: 954111861.65 / Avg: 955777557.21 / Max: 957932234.64 Min: 952998812.84 / Avg: 953591097.7 / Max: 953888243.32 Min: 950359782.15 / Avg: 954203726.6 / Max: 960169700.6 Min: 953952709.91 / Avg: 956651721.58 / Max: 959556588.76 Min: 951619924.77 / Avg: 956415576.33 / Max: 962491804.92 Min: 955930893.16 / Avg: 958824023.85 / Max: 962509279.98 Min: 950544598.53 / Avg: 953587374.96 / Max: 957321677.03 Min: 964947524.97 / Avg: 966044516.23 / Max: 967935126.49 1. (CC) gcc options: -march=native -pipe -O3 -lm
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 SE +/- 12.62, N = 3 SE +/- 2.80, N = 3 SE +/- 11.18, N = 3 SE +/- 9.68, N = 3 SE +/- 2.79, N = 3 SE +/- 29.99, N = 3 SE +/- 4.84, N = 3 SE +/- 0.00, N = 3 2397 2383 2403 2381 2395 2376 2389 2406 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 2372.72 / Avg: 2397.49 / Max: 2414.07 Min: 2380.54 / Avg: 2383.34 / Max: 2388.93 Min: 2380.54 / Avg: 2402.89 / Max: 2414.07 Min: 2363.78 / Avg: 2380.54 / Max: 2397.31 Min: 2388.93 / Avg: 2394.52 / Max: 2397.31 Min: 2315.72 / Avg: 2375.7 / Max: 2405.69 Min: 2380.54 / Avg: 2388.93 / Max: 2397.31 Min: 2405.69 / Avg: 2405.69 / Max: 2405.69
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Thread GRAPHITE LTO LTO GRAPHITE 0.9045 1.809 2.7135 3.618 4.5225 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.97 4.00 4.00 3.97 4.00 3.99 3.99 4.02 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Thread GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.97 / Avg: 3.97 / Max: 3.98 Min: 3.99 / Avg: 4 / Max: 4.01 Min: 3.98 / Avg: 4 / Max: 4.03 Min: 3.96 / Avg: 3.97 / Max: 3.97 Min: 3.98 / Avg: 4 / Max: 4.02 Min: 3.94 / Avg: 3.99 / Max: 4.05 Min: 3.98 / Avg: 3.99 / Max: 4 Min: 3.99 / Avg: 4.02 / Max: 4.04 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 492.06, N = 3 SE +/- 670.56, N = 3 SE +/- 195.52, N = 3 SE +/- 266.45, N = 3 SE +/- 259.53, N = 3 SE +/- 322.13, N = 3 SE +/- 364.54, N = 3 SE +/- 407.01, N = 3 51435 51605 51648 51520 51605 52032 51605 52075 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 50880 / Avg: 51434.67 / Max: 52416 Min: 50368 / Avg: 51605.33 / Max: 52672 Min: 51392 / Avg: 51648 / Max: 52032 Min: 51008 / Avg: 51520 / Max: 51904 Min: 51136 / Avg: 51605.33 / Max: 52032 Min: 51648 / Avg: 52032 / Max: 52672 Min: 50880 / Avg: 51605.33 / Max: 52032 Min: 51264 / Avg: 52074.67 / Max: 52544 1. (CC) gcc options: -O2 -lpthread
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 6.59 6.61 6.58 6.59 6.54 6.53 6.56 6.53 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 6.54 / Avg: 6.59 / Max: 6.64 Min: 6.58 / Avg: 6.61 / Max: 6.64 Min: 6.53 / Avg: 6.58 / Max: 6.62 Min: 6.58 / Avg: 6.59 / Max: 6.6 Min: 6.53 / Avg: 6.54 / Max: 6.55 Min: 6.51 / Avg: 6.53 / Max: 6.56 Min: 6.5 / Avg: 6.56 / Max: 6.66 Min: 6.48 / Avg: 6.53 / Max: 6.56 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Decrypt GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 2.83, N = 3 SE +/- 2.53, N = 3 SE +/- 4.34, N = 3 SE +/- 7.03, N = 3 SE +/- 1.71, N = 3 SE +/- 3.75, N = 3 SE +/- 1.27, N = 3 SE +/- 2.87, N = 3 3891 3879 3882 3881 3846 3846 3870 3844 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Decrypt GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 3887.67 / Avg: 3891.2 / Max: 3896.8 Min: 3874.44 / Avg: 3879.27 / Max: 3882.98 Min: 3873.75 / Avg: 3881.99 / Max: 3888.47 Min: 3867.22 / Avg: 3880.84 / Max: 3890.7 Min: 3843.19 / Avg: 3845.56 / Max: 3848.88 Min: 3841.16 / Avg: 3845.55 / Max: 3853.02 Min: 3867.54 / Avg: 3870.07 / Max: 3871.39 Min: 3838.93 / Avg: 3843.56 / Max: 3848.81 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 15K 30K 45K 60K 75K SE +/- 143.75, N = 3 SE +/- 162.24, N = 3 SE +/- 255.57, N = 3 SE +/- 461.97, N = 3 SE +/- 258.22, N = 3 SE +/- 194.23, N = 3 SE +/- 408.21, N = 3 SE +/- 129.21, N = 3 69307 68857 68642 69468 69245 68935 69130 69279 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K Min: 69065.53 / Avg: 69306.86 / Max: 69562.86 Min: 68641.91 / Avg: 68857.3 / Max: 69175.17 Min: 68253.02 / Avg: 68642.28 / Max: 69123.8 Min: 68545.42 / Avg: 69468.37 / Max: 69966.86 Min: 68877.38 / Avg: 69244.91 / Max: 69742.89 Min: 68550.01 / Avg: 68934.89 / Max: 69172.83 Min: 68360.09 / Avg: 69130.19 / Max: 69750.02 Min: 69053.88 / Avg: 69278.61 / Max: 69501.48 1. (CC) gcc options: -O2
Perl Benchmarks Perl benchmark suite that can be used to compare the relative speed of different versions of perl. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html GRAPHITE LTO LTO GRAPHITE 0.0238 0.0476 0.0714 0.0952 0.119 SE +/- 0.00025085, N = 3 SE +/- 0.00007060, N = 3 SE +/- 0.00012571, N = 3 SE +/- 0.00043538, N = 3 SE +/- 0.00026637, N = 3 SE +/- 0.00009383, N = 3 SE +/- 0.00015272, N = 3 SE +/- 0.00035883, N = 3 0.10510142 0.10460899 0.10573819 0.10529437 0.10506050 0.10450517 0.10505038 0.10448459 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html GRAPHITE LTO LTO GRAPHITE 1 2 3 4 5 Min: 0.1 / Avg: 0.11 / Max: 0.11 Min: 0.1 / Avg: 0.1 / Max: 0.1 Min: 0.11 / Avg: 0.11 / Max: 0.11 Min: 0.1 / Avg: 0.11 / Max: 0.11 Min: 0.1 / Avg: 0.11 / Max: 0.11 Min: 0.1 / Avg: 0.1 / Max: 0.1 Min: 0.1 / Avg: 0.11 / Max: 0.11 Min: 0.1 / Avg: 0.1 / Max: 0.1
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 641.02, N = 3 SE +/- 590.79, N = 3 SE +/- 437.24, N = 3 SE +/- 845.01, N = 3 SE +/- 442.37, N = 3 SE +/- 937.15, N = 3 SE +/- 331.84, N = 3 SE +/- 643.71, N = 3 138824 138419 138320 139596 137945 138018 138449 138458 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 137602.04 / Avg: 138823.59 / Max: 139771.37 Min: 137279.22 / Avg: 138419.29 / Max: 139258.15 Min: 137578.93 / Avg: 138320.39 / Max: 139092.64 Min: 138093.79 / Avg: 139595.84 / Max: 141017.7 Min: 137145.92 / Avg: 137944.71 / Max: 138673.53 Min: 136224.56 / Avg: 138018.17 / Max: 139386.1 Min: 137824.29 / Avg: 138448.57 / Max: 138955.8 Min: 137366.69 / Avg: 138457.64 / Max: 139595.12 1. (CC) gcc options: -O2
Hierarchical INTegration This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT GRAPHITE LTO LTO GRAPHITE 90M 180M 270M 360M 450M SE +/- 177359.32, N = 3 SE +/- 143994.60, N = 3 SE +/- 740071.73, N = 3 SE +/- 660332.35, N = 3 SE +/- 969130.90, N = 3 SE +/- 551831.61, N = 3 SE +/- 268239.00, N = 3 SE +/- 359623.43, N = 3 433045250 432962067 435372218 431821884 433258611 432529026 434699460 436963012 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT GRAPHITE LTO LTO GRAPHITE 80M 160M 240M 320M 400M Min: 432834509.15 / Avg: 433045249.61 / Max: 433397723.71 Min: 432707566.99 / Avg: 432962067.35 / Max: 433206045.89 Min: 434609700.67 / Avg: 435372217.51 / Max: 436852131.72 Min: 430798995.86 / Avg: 431821883.92 / Max: 433056785.92 Min: 431922709.33 / Avg: 433258611.15 / Max: 435142772.92 Min: 431443853.42 / Avg: 432529026.48 / Max: 433245836.9 Min: 434316680.26 / Avg: 434699459.94 / Max: 435216374.37 Min: 436408020.18 / Avg: 436963012.15 / Max: 437636708.81 1. (CC) gcc options: -march=native -pipe -O3 -lm
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time GRAPHITE LTO LTO GRAPHITE 13M 26M 39M 52M 65M SE +/- 317798.82, N = 3 SE +/- 419940.18, N = 3 SE +/- 163171.06, N = 3 SE +/- 442251.22, N = 3 SE +/- 335559.98, N = 3 SE +/- 544476.89, N = 3 SE +/- 207209.46, N = 3 SE +/- 112930.23, N = 3 57900023 57929115 58544928 57941895 58418663 58192490 57981415 57857797 1. (CXX) g++ options: -m64 -lpthread -march=native -pipe -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto Result Confidence
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time GRAPHITE LTO LTO GRAPHITE 10M 20M 30M 40M 50M Min: 57268616 / Avg: 57900023 / Max: 58278831 Min: 57395914 / Avg: 57929115.33 / Max: 58757696 Min: 58269089 / Avg: 58544927.67 / Max: 58833874 Min: 57073995 / Avg: 57941895 / Max: 58523563 Min: 58070560 / Avg: 58418663.33 / Max: 59089625 Min: 57405763 / Avg: 58192490 / Max: 59237897 Min: 57569324 / Avg: 57981414.67 / Max: 58225450 Min: 57720490 / Avg: 57857797 / Max: 58081756 1. (CXX) g++ options: -m64 -lpthread -march=native -pipe -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 10.53 10.43 10.42 10.47 10.45 10.54 10.42 10.46 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10.31 / Avg: 10.53 / Max: 10.86 Min: 10.39 / Avg: 10.43 / Max: 10.47 Min: 10.37 / Avg: 10.42 / Max: 10.46 Min: 10.41 / Avg: 10.47 / Max: 10.58 Min: 10.42 / Avg: 10.45 / Max: 10.48 Min: 10.52 / Avg: 10.54 / Max: 10.57 Min: 10.38 / Avg: 10.42 / Max: 10.49 Min: 10.27 / Avg: 10.46 / Max: 10.56 1. (CXX) g++ options: -O2 -lOpenCL
Stream This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 16.88, N = 5 SE +/- 11.15, N = 5 SE +/- 19.76, N = 5 SE +/- 11.26, N = 5 SE +/- 6.12, N = 5 SE +/- 10.25, N = 5 SE +/- 8.32, N = 5 SE +/- 8.18, N = 5 52285 52295 52701 52175 52111 52280 52103 52256 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 52232.9 / Avg: 52285.28 / Max: 52327 Min: 52268.7 / Avg: 52295.1 / Max: 52328.6 Min: 52647.2 / Avg: 52701.4 / Max: 52767.3 Min: 52147.7 / Avg: 52175.36 / Max: 52214.2 Min: 52090.2 / Avg: 52110.52 / Max: 52125.8 Min: 52260.2 / Avg: 52280.46 / Max: 52318.4 Min: 52086.6 / Avg: 52102.92 / Max: 52123.8 Min: 52236.6 / Avg: 52256.42 / Max: 52284.2 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp
Fhourstones This integer benchmark solves positions in the game of Connect-4, as played on a vertical 7x6 board. By default, it uses a 64Mb transposition table with the twobig replacement strategy. Positions are represented as 64-bit bitboards, and the hash function is computed using a single 64-bit modulo operation, giving 64-bit machines a slight edge. The alpha-beta searcher sorts moves dynamically based on the history heuristic. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Kpos / sec, More Is Better Fhourstones 3.1 Complex Connect-4 Solving GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 12.01, N = 3 SE +/- 9.47, N = 3 SE +/- 18.55, N = 3 SE +/- 7.46, N = 3 SE +/- 9.79, N = 3 SE +/- 1.96, N = 3 SE +/- 19.59, N = 3 SE +/- 4.02, N = 3 15247 15158 15263 15090 15241 15263 15232 15240 1. (CC) gcc options: -O3 Result Confidence
O2 O3 OpenBenchmarking.org Kpos / sec, More Is Better Fhourstones 3.1 Complex Connect-4 Solving GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 15226 / Avg: 15246.93 / Max: 15267.6 Min: 15145.7 / Avg: 15157.97 / Max: 15176.6 Min: 15225.7 / Avg: 15262.67 / Max: 15283.9 Min: 15075.3 / Avg: 15090.1 / Max: 15099.2 Min: 15224.3 / Avg: 15240.93 / Max: 15258.2 Min: 15259.3 / Avg: 15262.73 / Max: 15266.1 Min: 15193.1 / Avg: 15231.57 / Max: 15257.2 Min: 15234.8 / Avg: 15239.5 / Max: 15247.5 1. (CC) gcc options: -O3
GnuPG This test times how long it takes to encrypt a file using GnuPG. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better GnuPG 1.4.22 2GB File Encryption GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 6.20 6.19 6.18 6.20 6.18 6.20 6.13 6.17 1. (CC) gcc options: -march=native -pipe -O3 -MT -MD -MP -MF Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better GnuPG 1.4.22 2GB File Encryption GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 6.18 / Avg: 6.2 / Max: 6.23 Min: 6.18 / Avg: 6.19 / Max: 6.21 Min: 6.17 / Avg: 6.18 / Max: 6.2 Min: 6.17 / Avg: 6.2 / Max: 6.24 Min: 6.16 / Avg: 6.18 / Max: 6.22 Min: 6.16 / Avg: 6.2 / Max: 6.29 Min: 6.11 / Avg: 6.13 / Max: 6.16 Min: 6.12 / Avg: 6.17 / Max: 6.25 1. (CC) gcc options: -march=native -pipe -O3 -MT -MD -MP -MF
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 89.43, N = 3 SE +/- 38.42, N = 3 SE +/- 11.08, N = 3 SE +/- 26.02, N = 3 SE +/- 30.81, N = 3 SE +/- 56.18, N = 3 SE +/- 81.93, N = 3 SE +/- 67.51, N = 3 32298 32176 32152 32335 32451 32514 32356 32399 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 32150.71 / Avg: 32297.9 / Max: 32459.51 Min: 32102.33 / Avg: 32176.05 / Max: 32231.67 Min: 32135.57 / Avg: 32151.72 / Max: 32172.93 Min: 32282.47 / Avg: 32334.5 / Max: 32361.03 Min: 32406.09 / Avg: 32450.99 / Max: 32510 Min: 32439.97 / Avg: 32513.64 / Max: 32623.93 Min: 32195.18 / Avg: 32356.33 / Max: 32462.59 Min: 32265.18 / Avg: 32398.85 / Max: 32482.16 1. (CC) gcc options: -O2
Loopback TCP Network Performance This test measures the loopback network adapter performance using a micro-benchmark to measure the TCP performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Loopback TCP Network Performance Time To Transfer 10GB Via Loopback GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 10.81 10.85 10.76 10.73 10.79 10.78 10.85 10.80 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Loopback TCP Network Performance Time To Transfer 10GB Via Loopback GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10.74 / Avg: 10.81 / Max: 10.84 Min: 10.84 / Avg: 10.85 / Max: 10.86 Min: 10.73 / Avg: 10.76 / Max: 10.8 Min: 10.71 / Avg: 10.73 / Max: 10.74 Min: 10.75 / Avg: 10.79 / Max: 10.86 Min: 10.75 / Avg: 10.78 / Max: 10.83 Min: 10.77 / Avg: 10.85 / Max: 10.95 Min: 10.79 / Avg: 10.8 / Max: 10.82
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 3.09, N = 4 SE +/- 10.65, N = 4 SE +/- 6.38, N = 4 SE +/- 6.31, N = 4 SE +/- 4.27, N = 4 SE +/- 6.76, N = 4 SE +/- 18.78, N = 4 SE +/- 20.26, N = 4 2839 2854 2846 2858 2849 2837 2861 2868 Result Confidence
O2 O3 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 Min: 2832 / Avg: 2838.75 / Max: 2847 Min: 2838 / Avg: 2853.5 / Max: 2885 Min: 2835 / Avg: 2845.5 / Max: 2864 Min: 2841 / Avg: 2858 / Max: 2869 Min: 2837 / Avg: 2849.25 / Max: 2856 Min: 2828 / Avg: 2837.25 / Max: 2857 Min: 2824 / Avg: 2861 / Max: 2910 Min: 2842 / Avg: 2868 / Max: 2928
t-test1 This is a test of t-test1 for basic memory allocator benchmarks. Note this test profile is currently very basic and the overall time does include the warmup time of the custom t-test1 compilation. Improvements welcome. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 GRAPHITE LTO LTO GRAPHITE 1.2578 2.5156 3.7734 5.0312 6.289 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 5.55 5.53 5.57 5.57 5.57 5.55 5.59 5.54 1. (CC) gcc options: -pthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 5.51 / Avg: 5.55 / Max: 5.59 Min: 5.52 / Avg: 5.53 / Max: 5.53 Min: 5.53 / Avg: 5.57 / Max: 5.62 Min: 5.54 / Avg: 5.57 / Max: 5.61 Min: 5.54 / Avg: 5.57 / Max: 5.62 Min: 5.51 / Avg: 5.55 / Max: 5.63 Min: 5.52 / Avg: 5.59 / Max: 5.64 Min: 5.51 / Avg: 5.54 / Max: 5.58 1. (CC) gcc options: -pthread -march=native -pipe -O3
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 41.26, N = 3 SE +/- 192.65, N = 3 SE +/- 40.95, N = 3 SE +/- 246.61, N = 3 SE +/- 115.50, N = 3 SE +/- 112.40, N = 3 SE +/- 28.34, N = 3 SE +/- 131.93, N = 3 30791 30493 30692 30461 30652 30747 30739 30532 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 128 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K Min: 30715 / Avg: 30790.67 / Max: 30857 Min: 30124 / Avg: 30492.67 / Max: 30774 Min: 30641 / Avg: 30692 / Max: 30773 Min: 30155 / Avg: 30461 / Max: 30949 Min: 30536 / Avg: 30652 / Max: 30883 Min: 30622 / Avg: 30746.67 / Max: 30971 Min: 30692 / Avg: 30739.33 / Max: 30790 Min: 30303 / Avg: 30532.33 / Max: 30760 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 13.32 13.30 13.44 13.42 13.39 13.34 13.30 13.30 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 13.3 / Avg: 13.32 / Max: 13.36 Min: 13.3 / Avg: 13.3 / Max: 13.3 Min: 13.29 / Avg: 13.44 / Max: 13.52 Min: 13.36 / Avg: 13.42 / Max: 13.53 Min: 13.31 / Avg: 13.39 / Max: 13.54 Min: 13.31 / Avg: 13.34 / Max: 13.38 Min: 13.3 / Avg: 13.3 / Max: 13.3 Min: 13.29 / Avg: 13.3 / Max: 13.3 1. (CXX) g++ options: -O2 -lOpenCL
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Single-Precision Float GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 2.06, N = 3 SE +/- 8.24, N = 3 SE +/- 6.14, N = 3 SE +/- 12.71, N = 3 SE +/- 1.69, N = 3 SE +/- 12.20, N = 3 SE +/- 4.18, N = 3 SE +/- 13.35, N = 3 3617 3649 3614 3638 3618 3636 3612 3629 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Single-Precision Float GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3613.85 / Avg: 3616.94 / Max: 3620.85 Min: 3632.39 / Avg: 3648.83 / Max: 3657.96 Min: 3601.7 / Avg: 3613.88 / Max: 3621.31 Min: 3613.37 / Avg: 3638.19 / Max: 3655.32 Min: 3614.51 / Avg: 3617.81 / Max: 3620.12 Min: 3616.41 / Avg: 3635.94 / Max: 3658.38 Min: 3603.92 / Avg: 3612.29 / Max: 3616.56 Min: 3613.85 / Avg: 3628.58 / Max: 3655.24 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 15K 30K 45K 60K 75K SE +/- 107.37, N = 3 SE +/- 135.37, N = 3 SE +/- 9.22, N = 3 SE +/- 97.03, N = 3 SE +/- 116.46, N = 3 SE +/- 721.73, N = 3 SE +/- 190.12, N = 3 SE +/- 212.57, N = 3 67994 67630 67887 67727 67728 67312 67506 67822 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K Min: 67819.06 / Avg: 67994.27 / Max: 68189.39 Min: 67359.56 / Avg: 67629.99 / Max: 67776.42 Min: 67873 / Avg: 67887.25 / Max: 67904.51 Min: 67577.38 / Avg: 67727.39 / Max: 67909.01 Min: 67530.58 / Avg: 67728.25 / Max: 67933.79 Min: 65970.88 / Avg: 67312.39 / Max: 68444.62 Min: 67125.53 / Avg: 67505.69 / Max: 67702.48 Min: 67419.43 / Avg: 67821.89 / Max: 68141.74 1. (CC) gcc options: -O2
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.19, N = 3 SE +/- 0.24, N = 3 SE +/- 0.29, N = 3 SE +/- 0.04, N = 3 SE +/- 0.23, N = 3 SE +/- 0.20, N = 3 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 101 101 101 101 101 101 100 100 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 100.36 / Avg: 100.6 / Max: 100.98 Min: 100.12 / Avg: 100.52 / Max: 100.94 Min: 100.36 / Avg: 100.78 / Max: 101.34 Min: 100.5 / Avg: 100.57 / Max: 100.64 Min: 100.36 / Avg: 100.77 / Max: 101.15 Min: 100.41 / Avg: 100.74 / Max: 101.11 Min: 100.28 / Avg: 100.43 / Max: 100.68 Min: 100.43 / Avg: 100.45 / Max: 100.48 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 56.02, N = 3 SE +/- 43.52, N = 3 SE +/- 81.29, N = 3 SE +/- 136.59, N = 3 SE +/- 29.99, N = 3 SE +/- 121.45, N = 3 SE +/- 130.95, N = 3 SE +/- 93.24, N = 3 17212 17319 17364 17354 17258 17320 17313 17384 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 17155.88 / Avg: 17212.18 / Max: 17324.22 Min: 17265.8 / Avg: 17318.58 / Max: 17404.9 Min: 17251.26 / Avg: 17363.75 / Max: 17521.65 Min: 17083.75 / Avg: 17353.78 / Max: 17524.64 Min: 17201.99 / Avg: 17257.56 / Max: 17304.9 Min: 17079.76 / Avg: 17320.47 / Max: 17469.04 Min: 17051.32 / Avg: 17313.13 / Max: 17449.99 Min: 17197.94 / Avg: 17384.24 / Max: 17484.55 1. (CC) gcc options: -O2
lzbench lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Compression GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 SE +/- 4.84, N = 3 SE +/- 4.91, N = 3 SE +/- 5.67, N = 3 SE +/- 3.51, N = 3 SE +/- 4.37, N = 3 SE +/- 5.33, N = 3 SE +/- 5.36, N = 3 SE +/- 5.17, N = 3 417 417 417 417 416 415 415 419 1. (CXX) g++ options: -lrt -static -lpthread Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Compression GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 412 / Avg: 417.33 / Max: 427 Min: 411 / Avg: 417.33 / Max: 427 Min: 411 / Avg: 416.67 / Max: 428 Min: 413 / Avg: 417 / Max: 424 Min: 411 / Avg: 416.33 / Max: 425 Min: 410 / Avg: 415.33 / Max: 426 Min: 409 / Avg: 415.33 / Max: 426 Min: 413 / Avg: 418.67 / Max: 429 1. (CXX) g++ options: -lrt -static -lpthread
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 22.91, N = 3 SE +/- 15.80, N = 3 SE +/- 13.50, N = 3 SE +/- 23.68, N = 3 SE +/- 14.13, N = 3 SE +/- 18.71, N = 3 SE +/- 21.18, N = 3 SE +/- 12.42, N = 3 16320 16280 16261 16381 16416 16397 16329 16400 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 4 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 16275.93 / Avg: 16320.44 / Max: 16352.08 Min: 16254.74 / Avg: 16280.1 / Max: 16309.11 Min: 16240.05 / Avg: 16261.13 / Max: 16286.28 Min: 16350.25 / Avg: 16381.45 / Max: 16427.9 Min: 16394.23 / Avg: 16415.81 / Max: 16442.4 Min: 16377.97 / Avg: 16397.07 / Max: 16434.49 Min: 16289.13 / Avg: 16329.36 / Max: 16360.96 Min: 16375.62 / Avg: 16400.2 / Max: 16415.52 1. (CC) gcc options: -O2
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 151.24, N = 3 SE +/- 53.25, N = 3 SE +/- 41.63, N = 3 SE +/- 277.22, N = 3 SE +/- 37.18, N = 3 SE +/- 9.07, N = 3 SE +/- 144.08, N = 3 SE +/- 96.78, N = 3 18341 18317 18425 18380 18415 18252 18421 18276 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 18058 / Avg: 18341 / Max: 18575 Min: 18246 / Avg: 18316.67 / Max: 18421 Min: 18345 / Avg: 18425 / Max: 18485 Min: 17943 / Avg: 18380 / Max: 18894 Min: 18341 / Avg: 18415.33 / Max: 18454 Min: 18238 / Avg: 18252 / Max: 18269 Min: 18205 / Avg: 18420.67 / Max: 18694 Min: 18171 / Avg: 18275.67 / Max: 18469 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 15K 30K 45K 60K 75K SE +/- 195.90, N = 3 SE +/- 289.07, N = 3 SE +/- 284.72, N = 3 SE +/- 199.44, N = 3 SE +/- 181.06, N = 3 SE +/- 58.89, N = 3 SE +/- 254.78, N = 3 SE +/- 212.48, N = 3 69536 69427 69277 69210 69550 69199 68904 69304 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 128MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K Min: 69169.33 / Avg: 69535.62 / Max: 69839.21 Min: 68895.92 / Avg: 69426.92 / Max: 69890.46 Min: 68863.48 / Avg: 69276.56 / Max: 69822.55 Min: 68953.91 / Avg: 69210.21 / Max: 69603.05 Min: 69252.36 / Avg: 69549.62 / Max: 69877.34 Min: 69105.14 / Avg: 69198.64 / Max: 69307.43 Min: 68576.4 / Avg: 68904.36 / Max: 69406.08 Min: 68882.01 / Avg: 69304.05 / Max: 69558.13 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 46.33, N = 3 SE +/- 44.13, N = 3 SE +/- 168.01, N = 3 SE +/- 32.68, N = 3 SE +/- 68.11, N = 3 SE +/- 36.10, N = 3 SE +/- 69.53, N = 3 SE +/- 86.37, N = 3 31653 31454 31449 31647 31733 31583 31704 31607 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 31565.97 / Avg: 31652.62 / Max: 31724.39 Min: 31397.56 / Avg: 31453.62 / Max: 31540.69 Min: 31115.16 / Avg: 31448.86 / Max: 31649.87 Min: 31597.14 / Avg: 31647.33 / Max: 31708.68 Min: 31657.7 / Avg: 31732.55 / Max: 31868.54 Min: 31523.21 / Avg: 31582.93 / Max: 31647.92 Min: 31569.86 / Avg: 31703.75 / Max: 31803.22 Min: 31436.12 / Avg: 31607.36 / Max: 31712.61 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 30K 60K 90K 120K 150K SE +/- 122.13, N = 3 SE +/- 303.53, N = 3 SE +/- 413.24, N = 3 SE +/- 627.93, N = 3 SE +/- 449.68, N = 3 SE +/- 757.15, N = 3 SE +/- 902.54, N = 3 SE +/- 264.56, N = 3 136808 137732 137313 137183 137730 137689 137228 138042 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 20K 40K 60K 80K 100K Min: 136569.75 / Avg: 136807.93 / Max: 136973.93 Min: 137210.24 / Avg: 137732.16 / Max: 138261.6 Min: 136697.37 / Avg: 137313.15 / Max: 138098.45 Min: 136401.48 / Avg: 137183.03 / Max: 138425.14 Min: 136832.08 / Avg: 137729.91 / Max: 138224.28 Min: 136174.74 / Avg: 137689.05 / Max: 138446.2 Min: 135516.96 / Avg: 137227.52 / Max: 138582.04 Min: 137521.19 / Avg: 138042.06 / Max: 138383.05 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 15K 30K 45K 60K 75K SE +/- 143.72, N = 3 SE +/- 123.76, N = 3 SE +/- 80.05, N = 3 SE +/- 134.02, N = 3 SE +/- 154.89, N = 3 SE +/- 205.11, N = 3 SE +/- 81.64, N = 3 SE +/- 76.10, N = 3 71270 71335 70963 71213 71296 71391 71302 71577 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K Min: 70990.94 / Avg: 71270.07 / Max: 71469.08 Min: 71089.51 / Avg: 71335.08 / Max: 71484.67 Min: 70861.37 / Avg: 70963.04 / Max: 71120.99 Min: 70960.81 / Avg: 71213.41 / Max: 71417.36 Min: 70989.1 / Avg: 71295.59 / Max: 71487.79 Min: 71094.45 / Avg: 71390.95 / Max: 71784.72 Min: 71139.52 / Avg: 71302.14 / Max: 71396.2 Min: 71464.09 / Avg: 71577.06 / Max: 71721.87 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 135.58, N = 3 SE +/- 86.05, N = 3 SE +/- 209.66, N = 3 SE +/- 93.42, N = 3 SE +/- 129.73, N = 3 SE +/- 55.67, N = 3 SE +/- 42.40, N = 3 SE +/- 120.83, N = 3 34541 34517 34399 34519 34423 34346 34300 34586 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Write - Size Per Thread: 64MB - Thread Count: 32 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 34285.91 / Avg: 34541.4 / Max: 34747.79 Min: 34348.58 / Avg: 34516.7 / Max: 34632.62 Min: 33981.55 / Avg: 34399.42 / Max: 34638.48 Min: 34402.24 / Avg: 34518.91 / Max: 34703.63 Min: 34182.33 / Avg: 34422.88 / Max: 34627.35 Min: 34270.42 / Avg: 34346.46 / Max: 34454.91 Min: 34229.75 / Avg: 34299.98 / Max: 34376.26 Min: 34353.77 / Avg: 34585.9 / Max: 34760.18 1. (CC) gcc options: -O2
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Process GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.14, N = 3 SE +/- 0.10, N = 3 SE +/- 0.01, N = 3 43.40 43.59 43.44 43.23 43.43 43.39 43.34 43.39 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Process GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 Min: 43.21 / Avg: 43.4 / Max: 43.58 Min: 43.45 / Avg: 43.59 / Max: 43.69 Min: 43.28 / Avg: 43.44 / Max: 43.6 Min: 43.08 / Avg: 43.23 / Max: 43.36 Min: 43.35 / Avg: 43.43 / Max: 43.53 Min: 43.22 / Avg: 43.39 / Max: 43.67 Min: 43.15 / Avg: 43.34 / Max: 43.47 Min: 43.37 / Avg: 43.39 / Max: 43.4 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Bork File Encrypter Bork is a small, cross-platform file encryption utility. It is written in Java and designed to be included along with the files it encrypts for long-term storage. This test measures the amount of time it takes to encrypt a sample file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bork File Encrypter 1.4 File Encryption Time GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 8.56 8.58 8.61 8.55 8.54 8.57 8.55 8.57 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Bork File Encrypter 1.4 File Encryption Time GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 8.54 / Avg: 8.56 / Max: 8.59 Min: 8.56 / Avg: 8.58 / Max: 8.61 Min: 8.59 / Avg: 8.61 / Max: 8.64 Min: 8.54 / Avg: 8.55 / Max: 8.57 Min: 8.49 / Avg: 8.54 / Max: 8.58 Min: 8.53 / Avg: 8.57 / Max: 8.62 Min: 8.53 / Avg: 8.55 / Max: 8.57 Min: 8.54 / Avg: 8.57 / Max: 8.59
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Myocyte GRAPHITE LTO LTO GRAPHITE 11 22 33 44 55 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.15, N = 3 SE +/- 0.18, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.35, N = 3 46.87 47.09 47.05 47.04 47.06 46.95 46.95 47.25 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Myocyte GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 Min: 46.78 / Avg: 46.87 / Max: 47.05 Min: 47.03 / Avg: 47.09 / Max: 47.16 Min: 46.88 / Avg: 47.05 / Max: 47.35 Min: 46.73 / Avg: 47.04 / Max: 47.36 Min: 46.98 / Avg: 47.06 / Max: 47.13 Min: 46.87 / Avg: 46.95 / Max: 47.1 Min: 46.78 / Avg: 46.95 / Max: 47.1 Min: 46.8 / Avg: 47.25 / Max: 47.94 1. (CXX) g++ options: -O2 -lOpenCL
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode GRAPHITE LTO LTO GRAPHITE 80 160 240 320 400 SE +/- 5.69, N = 4 SE +/- 6.12, N = 3 SE +/- 7.33, N = 3 SE +/- 6.51, N = 3 SE +/- 6.59, N = 3 SE +/- 6.33, N = 3 SE +/- 7.20, N = 3 SE +/- 6.13, N = 4 379 380 378 378 379 381 380 381 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm Result Confidence
O2 O3 OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 361.66 / Avg: 378.69 / Max: 385.36 Min: 367.42 / Avg: 379.55 / Max: 387.1 Min: 363.42 / Avg: 378.05 / Max: 386.1 Min: 365.85 / Avg: 378.46 / Max: 387.6 Min: 366.08 / Avg: 379.1 / Max: 387.35 Min: 368.78 / Avg: 381.33 / Max: 389.11 Min: 366.3 / Avg: 380.35 / Max: 390.12 Min: 362.54 / Avg: 380.72 / Max: 389.36 1. (CC) gcc options: -march=native -pipe -O3 -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process GRAPHITE LTO LTO GRAPHITE 0.8685 1.737 2.6055 3.474 4.3425 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 3.83 3.86 3.84 3.85 3.84 3.83 3.84 3.83 1. (CC) gcc options: -lpthread -march=native -pipe -O3 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.82 / Avg: 3.83 / Max: 3.84 Min: 3.84 / Avg: 3.86 / Max: 3.87 Min: 3.82 / Avg: 3.84 / Max: 3.86 Min: 3.84 / Avg: 3.85 / Max: 3.85 Min: 3.79 / Avg: 3.84 / Max: 3.87 Min: 3.78 / Avg: 3.83 / Max: 3.87 Min: 3.82 / Avg: 3.84 / Max: 3.87 Min: 3.83 / Avg: 3.83 / Max: 3.85 1. (CC) gcc options: -lpthread -march=native -pipe -O3
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Encrypt GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 1.42, N = 3 SE +/- 1.54, N = 3 SE +/- 1.22, N = 3 SE +/- 1.47, N = 3 SE +/- 0.41, N = 3 SE +/- 3.02, N = 3 SE +/- 0.68, N = 3 SE +/- 4.46, N = 3 3862 3864 3859 3869 3855 3854 3839 3847 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Encrypt GRAPHITE LTO LTO GRAPHITE 700 1400 2100 2800 3500 Min: 3860.5 / Avg: 3862.34 / Max: 3865.13 Min: 3860.73 / Avg: 3863.8 / Max: 3865.38 Min: 3856.62 / Avg: 3859.06 / Max: 3860.32 Min: 3866.57 / Avg: 3868.79 / Max: 3871.58 Min: 3853.96 / Avg: 3854.77 / Max: 3855.3 Min: 3848.77 / Avg: 3853.8 / Max: 3859.21 Min: 3838.22 / Avg: 3838.91 / Max: 3840.28 Min: 3837.67 / Avg: 3846.59 / Max: 3851.3 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL BFS GRAPHITE LTO LTO GRAPHITE 0.2948 0.5896 0.8844 1.1792 1.474 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 1.30 1.30 1.31 1.31 1.30 1.31 1.30 1.31 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL BFS GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.28 / Avg: 1.3 / Max: 1.32 Min: 1.29 / Avg: 1.3 / Max: 1.31 Min: 1.31 / Avg: 1.31 / Max: 1.31 Min: 1.3 / Avg: 1.31 / Max: 1.32 Min: 1.29 / Avg: 1.3 / Max: 1.31 Min: 1.29 / Avg: 1.31 / Max: 1.32 Min: 1.29 / Avg: 1.3 / Max: 1.32 Min: 1.31 / Avg: 1.31 / Max: 1.32 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
WavPack Audio Encoding This test times how long it takes to encode a sample WAV file to WavPack format. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better WavPack Audio Encoding 5.1 WAV To WavPack GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 6.55 6.55 6.54 6.56 6.54 6.54 6.51 6.52 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better WavPack Audio Encoding 5.1 WAV To WavPack GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 6.55 / Avg: 6.55 / Max: 6.55 Min: 6.54 / Avg: 6.55 / Max: 6.56 Min: 6.54 / Avg: 6.54 / Max: 6.55 Min: 6.55 / Avg: 6.56 / Max: 6.56 Min: 6.54 / Avg: 6.54 / Max: 6.54 Min: 6.54 / Avg: 6.54 / Max: 6.55 Min: 6.51 / Avg: 6.51 / Max: 6.52 Min: 6.52 / Avg: 6.52 / Max: 6.52 1. (CC) gcc options: -march=native -pipe -O3 -lm
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Decrypt GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.20, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 398 398 398 398 396 397 399 397 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Decrypt GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 397.6 / Avg: 397.84 / Max: 397.97 Min: 397.62 / Avg: 397.78 / Max: 397.99 Min: 397.86 / Avg: 397.92 / Max: 398.04 Min: 397.65 / Avg: 397.72 / Max: 397.77 Min: 396.09 / Avg: 396.4 / Max: 396.6 Min: 397.13 / Avg: 397.34 / Max: 397.75 Min: 399.22 / Avg: 399.33 / Max: 399.44 Min: 396.94 / Avg: 396.96 / Max: 396.98 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Decrypt GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 0.03, N = 3 SE +/- 0.60, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 146 145 146 146 146 146 146 146 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Decrypt GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 145.52 / Avg: 145.58 / Max: 145.64 Min: 144.01 / Avg: 145.22 / Max: 145.82 Min: 145.6 / Avg: 145.7 / Max: 145.77 Min: 145.39 / Avg: 145.52 / Max: 145.61 Min: 145.49 / Avg: 145.56 / Max: 145.6 Min: 145.62 / Avg: 145.66 / Max: 145.69 Min: 145.65 / Avg: 145.74 / Max: 145.79 Min: 145.53 / Avg: 145.54 / Max: 145.58 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Encrypt GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 146 146 145 145 146 146 146 146 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Encrypt GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 145.61 / Avg: 145.69 / Max: 145.77 Min: 145.78 / Avg: 145.85 / Max: 145.88 Min: 145.32 / Avg: 145.45 / Max: 145.51 Min: 145.2 / Avg: 145.44 / Max: 145.57 Min: 145.59 / Avg: 145.62 / Max: 145.63 Min: 145.62 / Avg: 145.65 / Max: 145.69 Min: 145.5 / Avg: 145.66 / Max: 145.74 Min: 145.62 / Avg: 145.64 / Max: 145.67 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 13.26 13.24 13.33 13.32 13.26 13.24 13.32 13.32 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 13.25 / Avg: 13.26 / Max: 13.26 Min: 13.24 / Avg: 13.24 / Max: 13.25 Min: 13.33 / Avg: 13.33 / Max: 13.33 Min: 13.32 / Avg: 13.32 / Max: 13.32 Min: 13.26 / Avg: 13.26 / Max: 13.27 Min: 13.24 / Avg: 13.24 / Max: 13.25 Min: 13.31 / Avg: 13.32 / Max: 13.33 Min: 13.32 / Avg: 13.32 / Max: 13.32 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
Multichase Pointer Chaser This is a benchmark of Google's multichase pointer chaser program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 2 Threads GRAPHITE LTO LTO GRAPHITE 15 30 45 60 75 SE +/- 0.48, N = 3 SE +/- 0.24, N = 3 SE +/- 0.23, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.27, N = 3 SE +/- 0.21, N = 3 68.67 68.79 68.97 69.11 68.82 68.69 68.65 68.67 1. (CC) gcc options: -O2 -static -pthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 2 Threads GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 Min: 67.74 / Avg: 68.67 / Max: 69.32 Min: 68.41 / Avg: 68.79 / Max: 69.24 Min: 68.51 / Avg: 68.97 / Max: 69.25 Min: 68.94 / Avg: 69.11 / Max: 69.29 Min: 68.7 / Avg: 68.82 / Max: 68.9 Min: 68.39 / Avg: 68.69 / Max: 68.99 Min: 68.13 / Avg: 68.65 / Max: 69.02 Min: 68.38 / Avg: 68.67 / Max: 69.09 1. (CC) gcc options: -O2 -static -pthread -lrt
Smallpt Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples GRAPHITE LTO LTO GRAPHITE 1.0148 2.0296 3.0444 4.0592 5.074 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 4.50 4.49 4.50 4.50 4.51 4.48 4.51 4.50 1. (CXX) g++ options: -fopenmp -O3 -march=native -pipe Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 4.49 / Avg: 4.5 / Max: 4.5 Min: 4.47 / Avg: 4.49 / Max: 4.51 Min: 4.5 / Avg: 4.5 / Max: 4.51 Min: 4.49 / Avg: 4.5 / Max: 4.51 Min: 4.49 / Avg: 4.51 / Max: 4.53 Min: 4.46 / Avg: 4.48 / Max: 4.51 Min: 4.49 / Avg: 4.51 / Max: 4.55 Min: 4.47 / Avg: 4.5 / Max: 4.53 1. (CXX) g++ options: -fopenmp -O3 -march=native -pipe
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Heartwall GRAPHITE LTO LTO GRAPHITE 0.6795 1.359 2.0385 2.718 3.3975 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.00 3.02 3.00 3.00 3.02 3.01 3.02 3.01 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Heartwall GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3 / Avg: 3 / Max: 3.01 Min: 3.02 / Avg: 3.02 / Max: 3.04 Min: 3 / Avg: 3 / Max: 3.01 Min: 3 / Avg: 3 / Max: 3.02 Min: 3 / Avg: 3.02 / Max: 3.04 Min: 3 / Avg: 3.01 / Max: 3.04 Min: 3.01 / Avg: 3.02 / Max: 3.03 Min: 3 / Avg: 3.01 / Max: 3.02 1. (CXX) g++ options: -O2 -lOpenCL
Multichase Pointer Chaser This is a benchmark of Google's multichase pointer chaser program. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 4 Threads GRAPHITE LTO LTO GRAPHITE 16 32 48 64 80 SE +/- 0.12, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.21, N = 3 SE +/- 0.22, N = 3 SE +/- 0.03, N = 3 SE +/- 0.26, N = 3 SE +/- 0.39, N = 3 69.92 70.02 70.20 69.90 70.26 70.18 70.13 70.36 1. (CC) gcc options: -O2 -static -pthread -lrt Result Confidence
O2 O3 OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 4 Threads GRAPHITE LTO LTO GRAPHITE 14 28 42 56 70 Min: 69.73 / Avg: 69.92 / Max: 70.13 Min: 69.74 / Avg: 70.02 / Max: 70.33 Min: 70.06 / Avg: 70.2 / Max: 70.27 Min: 69.48 / Avg: 69.9 / Max: 70.12 Min: 69.82 / Avg: 70.26 / Max: 70.51 Min: 70.12 / Avg: 70.18 / Max: 70.23 Min: 69.73 / Avg: 70.13 / Max: 70.61 Min: 69.65 / Avg: 70.36 / Max: 71 1. (CC) gcc options: -O2 -static -pthread -lrt
Scikit-Learn Scikit-learn is a Python module for machine learning Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.17.1 GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 0.19, N = 3 SE +/- 0.16, N = 3 SE +/- 0.28, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 161 161 161 161 160 161 160 161 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.17.1 GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 160.41 / Avg: 160.66 / Max: 161.02 Min: 160.52 / Avg: 160.84 / Max: 161.07 Min: 160.2 / Avg: 160.7 / Max: 161.18 Min: 161.02 / Avg: 161.11 / Max: 161.21 Min: 160.26 / Avg: 160.3 / Max: 160.34 Min: 160.63 / Avg: 160.79 / Max: 160.92 Min: 159.94 / Avg: 160.11 / Max: 160.37 Min: 160.53 / Avg: 160.57 / Max: 160.64
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sin GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 SE +/- 0.21, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.14, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 44.90 44.65 44.70 44.68 44.92 44.68 44.67 44.68 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sin GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 Min: 44.69 / Avg: 44.9 / Max: 45.31 Min: 44.61 / Avg: 44.65 / Max: 44.68 Min: 44.66 / Avg: 44.7 / Max: 44.75 Min: 44.68 / Avg: 44.68 / Max: 44.69 Min: 44.7 / Avg: 44.92 / Max: 45.17 Min: 44.67 / Avg: 44.68 / Max: 44.69 Min: 44.65 / Avg: 44.67 / Max: 44.68 Min: 44.68 / Avg: 44.68 / Max: 44.69
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org GIOPS, More Is Better clpeak OpenCL Test: Integer Compute INT GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 0.13, N = 3 SE +/- 1.01, N = 3 SE +/- 0.61, N = 3 SE +/- 0.36, N = 3 SE +/- 1.76, N = 3 SE +/- 1.80, N = 3 SE +/- 0.62, N = 3 SE +/- 0.40, N = 3 1163 1167 1163 1168 1165 1164 1161 1162 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GIOPS, More Is Better clpeak OpenCL Test: Integer Compute INT GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1162.53 / Avg: 1162.75 / Max: 1162.97 Min: 1165.21 / Avg: 1166.98 / Max: 1168.71 Min: 1161.54 / Avg: 1162.76 / Max: 1163.41 Min: 1167.62 / Avg: 1168.26 / Max: 1168.88 Min: 1162.67 / Avg: 1164.72 / Max: 1168.22 Min: 1161.56 / Avg: 1163.53 / Max: 1167.12 Min: 1160.1 / Avg: 1161.27 / Max: 1162.23 Min: 1161.56 / Avg: 1162.3 / Max: 1162.92 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: build GRAPHITE LTO LTO GRAPHITE 4000M 8000M 12000M 16000M 20000M SE +/- 22727615.08, N = 3 SE +/- 76559453.96, N = 3 SE +/- 17799366.27, N = 3 SE +/- 47353765.29, N = 3 SE +/- 49239319.31, N = 3 SE +/- 21098888.19, N = 3 SE +/- 46754005.24, N = 3 SE +/- 36229303.38, N = 3 16677488403 16631712282 16618970574 16636354796 16629020672 16687163664 16649639286 16589010951 Result Confidence
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: build GRAPHITE LTO LTO GRAPHITE 3000M 6000M 9000M 12000M 15000M Min: 16653450624 / Avg: 16677488403.33 / Max: 16722917955 Min: 16490503411 / Avg: 16631712282.33 / Max: 16753591380 Min: 16593823647 / Avg: 16618970573.67 / Max: 16653365503 Min: 16560849917 / Avg: 16636354796 / Max: 16723618822 Min: 16539615810 / Avg: 16629020672.33 / Max: 16709480755 Min: 16651845066 / Avg: 16687163663.67 / Max: 16724821381 Min: 16577650901 / Avg: 16649639286.33 / Max: 16737315829 Min: 16525260619 / Avg: 16589010950.67 / Max: 16650712425
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffsll GRAPHITE LTO LTO GRAPHITE 0.3893 0.7786 1.1679 1.5572 1.9465 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.73 1.72 1.72 1.73 1.73 1.72 1.72 1.72 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffsll GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.72 / Avg: 1.72 / Max: 1.72
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffs GRAPHITE LTO LTO GRAPHITE 0.3893 0.7786 1.1679 1.5572 1.9465 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.73 1.72 1.72 1.73 1.73 1.73 1.72 1.72 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffs GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.72 / Avg: 1.72 / Max: 1.72 Min: 1.72 / Avg: 1.72 / Max: 1.72
Y-Cruncher Y-Cruncher is a multi-threaded Pi benchmark. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.7 Calculating 500M Pi Digits GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 10.62 10.59 10.58 10.61 10.62 10.63 10.63 10.64 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.7 Calculating 500M Pi Digits GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10.6 / Avg: 10.62 / Max: 10.63 Min: 10.59 / Avg: 10.59 / Max: 10.59 Min: 10.56 / Avg: 10.58 / Max: 10.6 Min: 10.61 / Avg: 10.61 / Max: 10.61 Min: 10.61 / Avg: 10.62 / Max: 10.63 Min: 10.62 / Avg: 10.63 / Max: 10.63 Min: 10.62 / Avg: 10.63 / Max: 10.64 Min: 10.63 / Avg: 10.64 / Max: 10.65
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL TPACF GRAPHITE LTO LTO GRAPHITE 0.405 0.81 1.215 1.62 2.025 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.80 1.80 1.79 1.80 1.80 1.80 1.80 1.80 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL TPACF GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.79 / Avg: 1.8 / Max: 1.81 Min: 1.8 / Avg: 1.8 / Max: 1.81 Min: 1.78 / Avg: 1.79 / Max: 1.79 Min: 1.79 / Avg: 1.8 / Max: 1.81 Min: 1.79 / Avg: 1.8 / Max: 1.8 Min: 1.79 / Avg: 1.8 / Max: 1.8 Min: 1.79 / Avg: 1.8 / Max: 1.81 Min: 1.79 / Avg: 1.8 / Max: 1.81 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Sample Pi Program A simple C++ program that calculates Pi to 8,765,4321 digits using the Leibniz formula. This test can be used for showcasing how to write a basic test profile. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sample Pi Program GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 9.06 9.06 9.06 9.11 9.06 9.06 9.06 9.06 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Sample Pi Program GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 9.05 / Avg: 9.06 / Max: 9.08 Min: 9.05 / Avg: 9.06 / Max: 9.08 Min: 9.06 / Avg: 9.06 / Max: 9.06 Min: 9.08 / Avg: 9.11 / Max: 9.16 Min: 9.05 / Avg: 9.06 / Max: 9.06 Min: 9.05 / Avg: 9.06 / Max: 9.06 Min: 9.05 / Avg: 9.06 / Max: 9.06 Min: 9.05 / Avg: 9.06 / Max: 9.06
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: json GRAPHITE LTO LTO GRAPHITE 500K 1000K 1500K 2000K 2500K SE +/- 7314.22, N = 3 SE +/- 463.46, N = 3 SE +/- 5459.73, N = 3 SE +/- 299.89, N = 3 SE +/- 4657.90, N = 3 SE +/- 3802.89, N = 3 SE +/- 662.06, N = 3 SE +/- 3267.29, N = 3 2338554 2331460 2334512 2344303 2337536 2335399 2336412 2342738 Result Confidence
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: json GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K Min: 2325856 / Avg: 2338554 / Max: 2351193 Min: 2330764 / Avg: 2331459.67 / Max: 2332338 Min: 2325070 / Avg: 2334511.67 / Max: 2343983 Min: 2343733 / Avg: 2344302.67 / Max: 2344750 Min: 2332836 / Avg: 2337536.33 / Max: 2346852 Min: 2328348 / Avg: 2335399 / Max: 2341394 Min: 2335124 / Avg: 2336412 / Max: 2337322 Min: 2336257 / Avg: 2342738.33 / Max: 2346700
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Monte Carlo GRAPHITE LTO LTO GRAPHITE 120 240 360 480 600 SE +/- 0.27, N = 3 SE +/- 0.38, N = 3 SE +/- 0.38, N = 3 SE +/- 0.79, N = 3 SE +/- 0.35, N = 3 SE +/- 0.46, N = 3 SE +/- 0.26, N = 3 SE +/- 0.36, N = 3 551 551 551 552 551 551 551 549 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Monte Carlo GRAPHITE LTO LTO GRAPHITE 100 200 300 400 500 Min: 550.09 / Avg: 550.57 / Max: 551.02 Min: 550.61 / Avg: 551.16 / Max: 551.88 Min: 550.36 / Avg: 550.95 / Max: 551.67 Min: 550.83 / Avg: 551.99 / Max: 553.49 Min: 550.28 / Avg: 550.98 / Max: 551.44 Min: 549.95 / Avg: 550.85 / Max: 551.44 Min: 550.11 / Avg: 550.5 / Max: 551 Min: 548.78 / Avg: 549.47 / Max: 549.96 1. (CC) gcc options: -lm -ldl
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP GRAPHITE LTO LTO GRAPHITE 0.432 0.864 1.296 1.728 2.16 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.92 1.92 1.91 1.91 1.92 1.92 1.92 1.91 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.92 / Avg: 1.92 / Max: 1.92 Min: 1.91 / Avg: 1.92 / Max: 1.92 Min: 1.91 / Avg: 1.91 / Max: 1.92 Min: 1.91 / Avg: 1.91 / Max: 1.92 Min: 1.91 / Avg: 1.92 / Max: 1.92 Min: 1.92 / Avg: 1.92 / Max: 1.92 Min: 1.92 / Avg: 1.92 / Max: 1.92 Min: 1.91 / Avg: 1.91 / Max: 1.92 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Stream This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 7.58, N = 5 SE +/- 3.93, N = 5 SE +/- 14.85, N = 5 SE +/- 2.97, N = 5 SE +/- 3.23, N = 5 SE +/- 3.16, N = 5 SE +/- 3.64, N = 5 SE +/- 4.10, N = 5 58091 58056 57792 57958 58035 58091 57934 57949 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 58065 / Avg: 58091.46 / Max: 58107.3 Min: 58040.9 / Avg: 58055.92 / Max: 58062 Min: 57736.7 / Avg: 57792.22 / Max: 57817.2 Min: 57947.4 / Avg: 57958.02 / Max: 57965.4 Min: 58027.2 / Avg: 58034.98 / Max: 58042.6 Min: 58084.5 / Avg: 58090.92 / Max: 58102.9 Min: 57920.7 / Avg: 57934.42 / Max: 57941.7 Min: 57936 / Avg: 57948.66 / Max: 57957.1 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: pthread_once GRAPHITE LTO LTO GRAPHITE 0.441 0.882 1.323 1.764 2.205 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.96 1.95 1.95 1.96 1.96 1.95 1.95 1.95 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: pthread_once GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 1.96 / Avg: 1.96 / Max: 1.96 Min: 1.95 / Avg: 1.95 / Max: 1.95 Min: 1.95 / Avg: 1.95 / Max: 1.95 Min: 1.96 / Avg: 1.96 / Max: 1.96 Min: 1.96 / Avg: 1.96 / Max: 1.96 Min: 1.95 / Avg: 1.95 / Max: 1.95 Min: 1.95 / Avg: 1.95 / Max: 1.95 Min: 1.95 / Avg: 1.95 / Max: 1.95
libjpeg-turbo tjbench tjbench is a JPEG decompression/compression benchmark part of libjpeg-turbo. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 SE +/- 0.69, N = 3 SE +/- 0.53, N = 3 SE +/- 0.27, N = 3 SE +/- 0.37, N = 3 SE +/- 0.14, N = 3 SE +/- 0.06, N = 3 SE +/- 0.59, N = 3 SE +/- 0.17, N = 3 197 197 197 197 196 197 197 197 1. (CC) gcc options: -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput GRAPHITE LTO LTO GRAPHITE 40 80 120 160 200 Min: 195.58 / Avg: 196.93 / Max: 197.87 Min: 195.69 / Avg: 196.62 / Max: 197.53 Min: 196.14 / Avg: 196.68 / Max: 196.99 Min: 196.72 / Avg: 197.37 / Max: 198.01 Min: 195.9 / Avg: 196.13 / Max: 196.37 Min: 196.97 / Avg: 197.07 / Max: 197.19 Min: 195.7 / Avg: 196.85 / Max: 197.69 Min: 196.6 / Avg: 196.93 / Max: 197.16 1. (CC) gcc options: -march=native -pipe -O3 -lm
Java Gradle Build This test runs Java software project builds using the Gradle build system. It is intended to give developers an idea as to the build performance for development activities and build servers. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Java Gradle Build 1.0 Gradle Build: Reactor GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 SE +/- 0.02, N = 4 SE +/- 0.05, N = 4 SE +/- 0.04, N = 4 SE +/- 0.05, N = 4 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 SE +/- 0.04, N = 4 19.71 19.67 19.63 19.69 19.68 19.70 19.73 19.66 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Java Gradle Build 1.0 Gradle Build: Reactor GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 Min: 19.66 / Avg: 19.71 / Max: 19.74 Min: 19.58 / Avg: 19.67 / Max: 19.81 Min: 19.58 / Avg: 19.63 / Max: 19.75 Min: 19.58 / Avg: 19.69 / Max: 19.8 Min: 19.6 / Avg: 19.68 / Max: 19.78 Min: 19.64 / Avg: 19.7 / Max: 19.76 Min: 19.64 / Avg: 19.73 / Max: 19.8 Min: 19.56 / Avg: 19.66 / Max: 19.71
Xsbench XSBench is a mini-app representing a key computational kernel of the Monte Carlo neutronics application OpenMC. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Lookups/s, More Is Better Xsbench 2017-07-06 GRAPHITE LTO LTO GRAPHITE 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 31414.50, N = 3 SE +/- 2711.20, N = 3 SE +/- 13313.57, N = 3 SE +/- 4067.06, N = 3 SE +/- 2732.01, N = 3 SE +/- 4253.63, N = 3 SE +/- 3099.86, N = 3 SE +/- 2087.75, N = 3 6025710 6041605 6023267 6029005 6030462 6049517 6044151 6053189 1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Lookups/s, More Is Better Xsbench 2017-07-06 GRAPHITE LTO LTO GRAPHITE 1000K 2000K 3000K 4000K 5000K Min: 5962894 / Avg: 6025709.67 / Max: 6058238 Min: 6037203 / Avg: 6041605 / Max: 6046548 Min: 6002774 / Avg: 6023267 / Max: 6048237 Min: 6021083 / Avg: 6029004.67 / Max: 6034565 Min: 6025800 / Avg: 6030462 / Max: 6035261 Min: 6041242 / Avg: 6049516.67 / Max: 6055365 Min: 6038047 / Avg: 6044150.67 / Max: 6048144 Min: 6049536 / Avg: 6053189.33 / Max: 6056767 1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -lm
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM GRAPHITE LTO LTO GRAPHITE 11 22 33 44 55 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 47.68 47.57 47.54 47.51 47.54 47.46 47.60 47.51 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 Min: 47.58 / Avg: 47.68 / Max: 47.74 Min: 47.53 / Avg: 47.57 / Max: 47.61 Min: 47.53 / Avg: 47.54 / Max: 47.56 Min: 47.49 / Avg: 47.51 / Max: 47.55 Min: 47.48 / Avg: 47.54 / Max: 47.65 Min: 47.41 / Avg: 47.46 / Max: 47.52 Min: 47.53 / Avg: 47.6 / Max: 47.65 Min: 47.43 / Avg: 47.51 / Max: 47.58 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance GRAPHITE LTO LTO GRAPHITE 1300 2600 3900 5200 6500 SE +/- 7.41, N = 3 SE +/- 9.33, N = 3 SE +/- 7.57, N = 3 SE +/- 2.74, N = 3 SE +/- 6.80, N = 3 SE +/- 1.62, N = 3 SE +/- 0.99, N = 3 SE +/- 9.05, N = 3 6052 6061 6075 6068 6056 6080 6054 6067 1. (CC) gcc options: -pthread -m64 -march=native -pipe -O3 -lssl -lcrypto -ldl Result Confidence
O2 O3 OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance GRAPHITE LTO LTO GRAPHITE 1100 2200 3300 4400 5500 Min: 6044 / Avg: 6052.43 / Max: 6067.2 Min: 6042.8 / Avg: 6060.57 / Max: 6074.4 Min: 6059.4 / Avg: 6074.53 / Max: 6082.4 Min: 6063.1 / Avg: 6068.37 / Max: 6072.3 Min: 6047.8 / Avg: 6055.67 / Max: 6069.2 Min: 6076.9 / Avg: 6080.1 / Max: 6082.1 Min: 6052.8 / Avg: 6054.4 / Max: 6056.2 Min: 6050.4 / Avg: 6066.63 / Max: 6081.7 1. (CC) gcc options: -pthread -m64 -march=native -pipe -O3 -lssl -lcrypto -ldl
O2 O3 OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Libdeflate 1 - Process: Decompression GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 1155 1155 1155 1150 1155 1155 1154 1155 1. (CXX) g++ options: -lrt -static -lpthread
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sqrt GRAPHITE LTO LTO GRAPHITE 0.5423 1.0846 1.6269 2.1692 2.7115 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.41 2.40 2.40 2.41 2.41 2.41 2.40 2.40 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sqrt GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 2.41 / Avg: 2.41 / Max: 2.41 Min: 2.4 / Avg: 2.4 / Max: 2.41 Min: 2.4 / Avg: 2.4 / Max: 2.4 Min: 2.41 / Avg: 2.41 / Max: 2.42 Min: 2.41 / Avg: 2.41 / Max: 2.41 Min: 2.4 / Avg: 2.41 / Max: 2.41 Min: 2.4 / Avg: 2.4 / Max: 2.4 Min: 2.4 / Avg: 2.4 / Max: 2.4
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: garbage GRAPHITE LTO LTO GRAPHITE 130K 260K 390K 520K 650K SE +/- 973.67, N = 3 SE +/- 1190.29, N = 3 SE +/- 1315.90, N = 3 SE +/- 1777.30, N = 3 SE +/- 987.83, N = 3 SE +/- 1544.80, N = 3 SE +/- 733.55, N = 3 SE +/- 1517.44, N = 3 621310 622956 623188 621051 623559 622412 621095 622652 Result Confidence
O2 O3 OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: garbage GRAPHITE LTO LTO GRAPHITE 110K 220K 330K 440K 550K Min: 619363 / Avg: 621310.33 / Max: 622285 Min: 620621 / Avg: 622956 / Max: 624525 Min: 620568 / Avg: 623188.33 / Max: 624711 Min: 618603 / Avg: 621051 / Max: 624507 Min: 621914 / Avg: 623558.67 / Max: 625329 Min: 620014 / Avg: 622411.67 / Max: 625298 Min: 619804 / Avg: 621095.33 / Max: 622344 Min: 621104 / Avg: 622652.33 / Max: 625687
NAMD CUDA NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. This version of the NAMD test profile uses CUDA GPU acceleration. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.13 ATPase Simulation - 327,506 Atoms GRAPHITE LTO LTO GRAPHITE 0.0764 0.1528 0.2292 0.3056 0.382 SE +/- 0.00090, N = 3 SE +/- 0.00152, N = 3 SE +/- 0.00063, N = 3 SE +/- 0.00066, N = 3 SE +/- 0.00077, N = 3 SE +/- 0.00054, N = 3 SE +/- 0.00077, N = 3 SE +/- 0.00081, N = 3 0.33925 0.33872 0.33808 0.33905 0.33893 0.33820 0.33944 0.33889 Result Confidence
O2 O3 OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.13 ATPase Simulation - 327,506 Atoms GRAPHITE LTO LTO GRAPHITE 1 2 3 4 5 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34 Min: 0.34 / Avg: 0.34 / Max: 0.34
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Particle Filter GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 12.61 12.60 12.60 12.63 12.60 12.60 12.65 12.61 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenCL Particle Filter GRAPHITE LTO LTO GRAPHITE 4 8 12 16 20 Min: 12.6 / Avg: 12.61 / Max: 12.63 Min: 12.58 / Avg: 12.6 / Max: 12.62 Min: 12.58 / Avg: 12.6 / Max: 12.62 Min: 12.61 / Avg: 12.63 / Max: 12.67 Min: 12.59 / Avg: 12.6 / Max: 12.62 Min: 12.55 / Avg: 12.6 / Max: 12.63 Min: 12.61 / Avg: 12.65 / Max: 12.68 Min: 12.57 / Avg: 12.61 / Max: 12.67 1. (CXX) g++ options: -O2 -lOpenCL
Crafty This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time GRAPHITE LTO LTO GRAPHITE 2M 4M 6M 8M 10M SE +/- 11858.51, N = 3 SE +/- 6818.47, N = 3 SE +/- 11085.94, N = 3 SE +/- 9502.97, N = 3 SE +/- 7504.06, N = 3 SE +/- 8579.23, N = 3 SE +/- 21898.78, N = 3 SE +/- 11653.10, N = 3 8848802 8816317 8823265 8832693 8816848 8817639 8829403 8816171 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm Result Confidence
O2 O3 OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time GRAPHITE LTO LTO GRAPHITE 1.5M 3M 4.5M 6M 7.5M Min: 8833141 / Avg: 8848801.67 / Max: 8872057 Min: 8809423 / Avg: 8816317.33 / Max: 8829954 Min: 8806892 / Avg: 8823265 / Max: 8844399 Min: 8821411 / Avg: 8832693 / Max: 8851580 Min: 8802877 / Avg: 8816848.33 / Max: 8828581 Min: 8801228 / Avg: 8817638.67 / Max: 8830183 Min: 8799100 / Avg: 8829403 / Max: 8871940 Min: 8794656 / Avg: 8816171.33 / Max: 8834688 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 2.17, N = 3 SE +/- 0.97, N = 3 SE +/- 2.04, N = 3 SE +/- 2.14, N = 3 SE +/- 1.44, N = 3 SE +/- 1.20, N = 3 SE +/- 1.27, N = 3 SE +/- 2.59, N = 3 1408 1411 1412 1407 1408 1409 1411 1408 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1404.53 / Avg: 1408.18 / Max: 1412.04 Min: 1409.74 / Avg: 1410.8 / Max: 1412.74 Min: 1408.81 / Avg: 1412.49 / Max: 1415.87 Min: 1404.72 / Avg: 1406.9 / Max: 1411.19 Min: 1405.19 / Avg: 1408.05 / Max: 1409.82 Min: 1406.67 / Avg: 1408.95 / Max: 1410.74 Min: 1409.13 / Avg: 1411.44 / Max: 1413.52 Min: 1402.72 / Avg: 1407.8 / Max: 1411.19 1. (CC) gcc options: -lm -ldl
Stream This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy GRAPHITE LTO LTO GRAPHITE 16K 32K 48K 64K 80K SE +/- 31.05, N = 5 SE +/- 36.79, N = 5 SE +/- 25.15, N = 5 SE +/- 58.51, N = 5 SE +/- 11.76, N = 5 SE +/- 22.04, N = 5 SE +/- 33.32, N = 5 SE +/- 22.42, N = 5 74769 74693 74557 74743 74755 74790 74720 74720 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K Min: 74679.1 / Avg: 74769 / Max: 74829 Min: 74596.1 / Avg: 74692.64 / Max: 74787.3 Min: 74484.3 / Avg: 74557.3 / Max: 74623.4 Min: 74581.1 / Avg: 74742.8 / Max: 74927.6 Min: 74720.7 / Avg: 74754.96 / Max: 74783.1 Min: 74707.3 / Avg: 74789.64 / Max: 74836.5 Min: 74640.9 / Avg: 74719.54 / Max: 74814.8 Min: 74651.7 / Avg: 74720.18 / Max: 74769.8 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Transfer Bandwidth enqueueReadBuffer GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 6.42 6.43 6.42 6.43 6.43 6.43 6.42 6.41 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Transfer Bandwidth enqueueReadBuffer GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 6.42 / Avg: 6.42 / Max: 6.42 Min: 6.42 / Avg: 6.43 / Max: 6.43 Min: 6.42 / Avg: 6.42 / Max: 6.42 Min: 6.42 / Avg: 6.43 / Max: 6.43 Min: 6.42 / Avg: 6.43 / Max: 6.43 Min: 6.42 / Avg: 6.43 / Max: 6.43 Min: 6.41 / Avg: 6.42 / Max: 6.43 Min: 6.41 / Avg: 6.41 / Max: 6.42 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Rust Prime Benchmark Based on petehunt/rust-benchmark, this is a prime number benchmark that is multi-threaded and written in Rustlang. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rust Prime Benchmark Prime Number Test To 200,000,000 GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.47 6.47 6.47 6.48 6.47 6.46 6.47 6.47 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rust Prime Benchmark Prime Number Test To 200,000,000 GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 6.46 / Avg: 6.47 / Max: 6.47 Min: 6.47 / Avg: 6.47 / Max: 6.47 Min: 6.46 / Avg: 6.47 / Max: 6.49 Min: 6.46 / Avg: 6.48 / Max: 6.49 Min: 6.46 / Avg: 6.47 / Max: 6.47 Min: 6.46 / Avg: 6.46 / Max: 6.46 Min: 6.47 / Avg: 6.47 / Max: 6.47 Min: 6.46 / Avg: 6.47 / Max: 6.47
Stream This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K SE +/- 7.62, N = 5 SE +/- 8.21, N = 5 SE +/- 11.46, N = 5 SE +/- 6.98, N = 5 SE +/- 11.12, N = 5 SE +/- 7.23, N = 5 SE +/- 7.11, N = 5 SE +/- 6.72, N = 5 58911 58892 58990 58902 58868 58926 58824 58919 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 58882.9 / Avg: 58911.04 / Max: 58927.4 Min: 58869.5 / Avg: 58892.2 / Max: 58910.1 Min: 58960.9 / Avg: 58989.84 / Max: 59030.4 Min: 58879.8 / Avg: 58902.34 / Max: 58920.1 Min: 58849.5 / Avg: 58868.26 / Max: 58911.5 Min: 58904.3 / Avg: 58925.68 / Max: 58946.4 Min: 58806.6 / Avg: 58823.62 / Max: 58842.3 Min: 58908.8 / Avg: 58918.92 / Max: 58945 1. (CC) gcc options: -march=native -pipe -O3 -fopenmp
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Atol GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 58.99 58.99 58.88 58.92 58.83 58.89 58.86 58.96 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Atol GRAPHITE LTO LTO GRAPHITE 12 24 36 48 60 Min: 58.87 / Avg: 58.99 / Max: 59.08 Min: 58.9 / Avg: 58.99 / Max: 59.12 Min: 58.85 / Avg: 58.88 / Max: 58.92 Min: 58.88 / Avg: 58.92 / Max: 58.98 Min: 58.82 / Avg: 58.83 / Max: 58.85 Min: 58.87 / Avg: 58.89 / Max: 58.91 Min: 58.85 / Avg: 58.86 / Max: 58.86 Min: 58.89 / Avg: 58.96 / Max: 59.03 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: cos GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.10, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 45.23 45.22 45.32 45.23 45.23 45.23 45.22 45.22 Result Confidence
O2 O3 OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: cos GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 Min: 45.22 / Avg: 45.23 / Max: 45.23 Min: 45.21 / Avg: 45.22 / Max: 45.22 Min: 45.22 / Avg: 45.32 / Max: 45.53 Min: 45.23 / Avg: 45.23 / Max: 45.23 Min: 45.22 / Avg: 45.23 / Max: 45.23 Min: 45.23 / Avg: 45.23 / Max: 45.23 Min: 45.22 / Avg: 45.22 / Max: 45.22 Min: 45.22 / Avg: 45.22 / Max: 45.22
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Abstraction GRAPHITE LTO LTO GRAPHITE 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 32.42 32.45 32.39 32.38 32.45 32.45 32.39 32.38 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Abstraction GRAPHITE LTO LTO GRAPHITE 7 14 21 28 35 Min: 32.41 / Avg: 32.42 / Max: 32.43 Min: 32.45 / Avg: 32.45 / Max: 32.45 Min: 32.39 / Avg: 32.39 / Max: 32.4 Min: 32.36 / Avg: 32.38 / Max: 32.41 Min: 32.45 / Avg: 32.45 / Max: 32.45 Min: 32.45 / Avg: 32.45 / Max: 32.45 Min: 32.38 / Avg: 32.39 / Max: 32.4 Min: 32.37 / Avg: 32.38 / Max: 32.38 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
Primesieve Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.4 1e12 Prime Number Generation GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 10.02 10.04 10.04 10.03 10.04 10.04 10.03 10.04 1. (CXX) g++ options: -march=native -pipe -O3 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.4 1e12 Prime Number Generation GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10 / Avg: 10.02 / Max: 10.05 Min: 10.03 / Avg: 10.04 / Max: 10.05 Min: 9.99 / Avg: 10.04 / Max: 10.06 Min: 10.01 / Avg: 10.03 / Max: 10.04 Min: 10.03 / Avg: 10.04 / Max: 10.06 Min: 10.03 / Avg: 10.04 / Max: 10.05 Min: 10 / Avg: 10.03 / Max: 10.07 Min: 10.01 / Avg: 10.04 / Max: 10.11 1. (CXX) g++ options: -march=native -pipe -O3 -lpthread
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 SE +/- 0.60, N = 4 SE +/- 0.51, N = 4 SE +/- 0.00, N = 4 SE +/- 0.31, N = 4 SE +/- 0.36, N = 4 SE +/- 0.00, N = 4 SE +/- 3.63, N = 4 SE +/- 0.00, N = 4 2558 2558 2558 2557 2558 2558 2553 2558 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Sparse Matrix Multiply GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 Min: 2556.26 / Avg: 2557.81 / Max: 2558.75 Min: 2556.26 / Avg: 2557.5 / Max: 2558.75 Min: 2557.5 / Avg: 2557.5 / Max: 2557.5 Min: 2556.26 / Avg: 2557.19 / Max: 2557.5 Min: 2557.5 / Avg: 2558.13 / Max: 2558.75 Min: 2557.5 / Avg: 2557.5 / Max: 2557.5 Min: 2542.62 / Avg: 2553.47 / Max: 2557.5 Min: 2557.5 / Avg: 2557.5 / Max: 2557.5
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL LBM GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 22.09 22.10 22.08 22.10 22.08 22.07 22.08 22.06 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenCL LBM GRAPHITE LTO LTO GRAPHITE 5 10 15 20 25 Min: 22.08 / Avg: 22.09 / Max: 22.1 Min: 22.09 / Avg: 22.1 / Max: 22.13 Min: 22.07 / Avg: 22.08 / Max: 22.11 Min: 22.09 / Avg: 22.1 / Max: 22.12 Min: 22.07 / Avg: 22.08 / Max: 22.09 Min: 22.06 / Avg: 22.07 / Max: 22.09 Min: 22.07 / Avg: 22.08 / Max: 22.09 Min: 22.04 / Avg: 22.06 / Max: 22.07 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms GRAPHITE LTO LTO GRAPHITE 0.2021 0.4042 0.6063 0.8084 1.0105 SE +/- 0.00068, N = 3 SE +/- 0.00098, N = 3 SE +/- 0.00128, N = 3 SE +/- 0.00042, N = 3 SE +/- 0.00087, N = 3 SE +/- 0.00118, N = 3 SE +/- 0.00086, N = 3 SE +/- 0.00122, N = 3 0.89727 0.89663 0.89695 0.89758 0.89795 0.89823 0.89663 0.89805 Result Confidence
O2 O3 OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 0.9 / Avg: 0.9 / Max: 0.9 Min: 0.89 / Avg: 0.9 / Max: 0.9 Min: 0.89 / Avg: 0.9 / Max: 0.9 Min: 0.9 / Avg: 0.9 / Max: 0.9 Min: 0.9 / Avg: 0.9 / Max: 0.9 Min: 0.9 / Avg: 0.9 / Max: 0.9 Min: 0.9 / Avg: 0.9 / Max: 0.9 Min: 0.9 / Avg: 0.9 / Max: 0.9
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Transfer Bandwidth enqueueWriteBuffer GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.11 6.10 6.10 6.11 6.10 6.10 6.10 6.11 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Transfer Bandwidth enqueueWriteBuffer GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 6.1 / Avg: 6.11 / Max: 6.12 Min: 6.1 / Avg: 6.1 / Max: 6.1 Min: 6.1 / Avg: 6.1 / Max: 6.1 Min: 6.1 / Avg: 6.11 / Max: 6.11 Min: 6.1 / Avg: 6.1 / Max: 6.1 Min: 6.1 / Avg: 6.1 / Max: 6.1 Min: 6.1 / Avg: 6.1 / Max: 6.1 Min: 6.11 / Avg: 6.11 / Max: 6.12 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Vector GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 83.09 83.09 83.14 83.16 83.08 83.10 83.19 83.16 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Vector GRAPHITE LTO LTO GRAPHITE 16 32 48 64 80 Min: 83.08 / Avg: 83.09 / Max: 83.11 Min: 83.07 / Avg: 83.09 / Max: 83.1 Min: 83.12 / Avg: 83.14 / Max: 83.15 Min: 83.14 / Avg: 83.16 / Max: 83.17 Min: 83.08 / Avg: 83.08 / Max: 83.09 Min: 83.09 / Avg: 83.1 / Max: 83.1 Min: 83.16 / Avg: 83.19 / Max: 83.24 Min: 83.15 / Avg: 83.16 / Max: 83.17 1. (CXX) g++ options: -march=native -pipe -O3 -std=c++11
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Decrypt GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 95.43 95.42 95.33 95.43 95.42 95.39 95.42 95.38 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Decrypt GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 95.42 / Avg: 95.43 / Max: 95.43 Min: 95.42 / Avg: 95.42 / Max: 95.43 Min: 95.3 / Avg: 95.33 / Max: 95.39 Min: 95.43 / Avg: 95.43 / Max: 95.44 Min: 95.4 / Avg: 95.42 / Max: 95.43 Min: 95.32 / Avg: 95.39 / Max: 95.42 Min: 95.41 / Avg: 95.42 / Max: 95.43 Min: 95.27 / Avg: 95.38 / Max: 95.44 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
CacheBench This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K SE +/- 1.36, N = 3 SE +/- 0.24, N = 3 SE +/- 22.81, N = 3 SE +/- 1.21, N = 3 SE +/- 0.29, N = 3 SE +/- 0.37, N = 3 SE +/- 0.28, N = 3 SE +/- 0.84, N = 3 31175 31177 31155 31177 31178 31179 31177 31178 1. (CC) gcc options: -lrt Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K Min: 31172.75 / Avg: 31175.47 / Max: 31176.89 Min: 31176.69 / Avg: 31177.14 / Max: 31177.53 Min: 31109.25 / Avg: 31154.86 / Max: 31178.06 Min: 31174.67 / Avg: 31176.98 / Max: 31178.77 Min: 31177.4 / Avg: 31177.73 / Max: 31178.32 Min: 31177.92 / Avg: 31178.65 / Max: 31179.15 Min: 31176.48 / Avg: 31176.97 / Max: 31177.44 Min: 31176.17 / Avg: 31177.83 / Max: 31178.74 1. (CC) gcc options: -lrt
Rust Mandelbrot This test profile is of the combined time for the serial and parallel Mandelbrot sets written in Rustlang via willi-kappler/mandel-rust. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot GRAPHITE LTO LTO GRAPHITE 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 41.96 41.96 41.97 41.95 41.98 41.95 41.96 41.95 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot GRAPHITE LTO LTO GRAPHITE 9 18 27 36 45 Min: 41.94 / Avg: 41.96 / Max: 41.97 Min: 41.94 / Avg: 41.96 / Max: 41.98 Min: 41.95 / Avg: 41.97 / Max: 41.99 Min: 41.94 / Avg: 41.95 / Max: 41.97 Min: 41.94 / Avg: 41.98 / Max: 42.01 Min: 41.94 / Avg: 41.95 / Max: 41.96 Min: 41.95 / Avg: 41.96 / Max: 41.98 Min: 41.94 / Avg: 41.95 / Max: 41.96
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 0.17, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.84, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 1603 1603 1603 1604 1603 1603 1603 1603 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 Min: 1603.12 / Avg: 1603.29 / Max: 1603.79 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12 Min: 1603.12 / Avg: 1603.96 / Max: 1606.47 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12 Min: 1603.12 / Avg: 1603.12 / Max: 1603.12
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 400 800 1200 1600 2000 SE +/- 0.18, N = 3 SE +/- 0.34, N = 3 SE +/- 0.06, N = 3 SE +/- 0.27, N = 3 SE +/- 0.47, N = 3 SE +/- 0.16, N = 3 SE +/- 0.59, N = 3 SE +/- 0.06, N = 3 1653 1654 1654 1653 1654 1654 1654 1654 1. (CC) gcc options: -lm -ldl Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 Min: 1653.09 / Avg: 1653.43 / Max: 1653.67 Min: 1653.82 / Avg: 1654.49 / Max: 1654.95 Min: 1654.41 / Avg: 1654.47 / Max: 1654.6 Min: 1652.43 / Avg: 1652.95 / Max: 1653.36 Min: 1652.81 / Avg: 1653.72 / Max: 1654.4 Min: 1653.92 / Avg: 1654.17 / Max: 1654.48 Min: 1652.71 / Avg: 1653.8 / Max: 1654.72 Min: 1653.72 / Avg: 1653.82 / Max: 1653.91 1. (CC) gcc options: -lm -ldl
CacheBench This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read GRAPHITE LTO LTO GRAPHITE 800 1600 2400 3200 4000 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 1.15, N = 3 SE +/- 0.22, N = 3 SE +/- 0.16, N = 3 SE +/- 0.23, N = 3 SE +/- 0.05, N = 3 SE +/- 0.23, N = 3 3726 3725 3724 3725 3725 3725 3725 3725 1. (CC) gcc options: -lrt Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 Min: 3725.64 / Avg: 3725.75 / Max: 3725.87 Min: 3725.3 / Avg: 3725.48 / Max: 3725.58 Min: 3721.81 / Avg: 3724.11 / Max: 3725.28 Min: 3724.23 / Avg: 3724.68 / Max: 3724.92 Min: 3724.78 / Avg: 3724.96 / Max: 3725.29 Min: 3724.48 / Avg: 3724.93 / Max: 3725.17 Min: 3724.59 / Avg: 3724.66 / Max: 3724.75 Min: 3724.6 / Avg: 3724.83 / Max: 3725.29 1. (CC) gcc options: -lrt
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Encrypt GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 96.54 96.50 96.49 96.53 96.52 96.52 96.52 96.52 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Encrypt GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 Min: 96.54 / Avg: 96.54 / Max: 96.54 Min: 96.49 / Avg: 96.5 / Max: 96.52 Min: 96.46 / Avg: 96.49 / Max: 96.53 Min: 96.53 / Avg: 96.53 / Max: 96.54 Min: 96.5 / Avg: 96.52 / Max: 96.54 Min: 96.49 / Avg: 96.52 / Max: 96.54 Min: 96.47 / Avg: 96.52 / Max: 96.54 Min: 96.48 / Avg: 96.52 / Max: 96.54 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
CacheBench This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read / Modify / Write GRAPHITE LTO LTO GRAPHITE 11K 22K 33K 44K 55K SE +/- 4.66, N = 3 SE +/- 4.03, N = 3 SE +/- 4.97, N = 3 SE +/- 2.40, N = 3 SE +/- 2.79, N = 3 SE +/- 10.30, N = 3 SE +/- 3.39, N = 3 SE +/- 1.57, N = 3 50803 50794 50793 50795 50800 50804 50793 50797 1. (CC) gcc options: -lrt Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read / Modify / Write GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 50794.28 / Avg: 50803.06 / Max: 50810.18 Min: 50785.96 / Avg: 50793.75 / Max: 50799.42 Min: 50785.1 / Avg: 50792.59 / Max: 50801.99 Min: 50791.99 / Avg: 50795.21 / Max: 50799.9 Min: 50796.72 / Avg: 50800.46 / Max: 50805.9 Min: 50789.54 / Avg: 50803.57 / Max: 50823.65 Min: 50787.77 / Avg: 50792.6 / Max: 50799.14 Min: 50794.32 / Avg: 50797.44 / Max: 50799.32 1. (CC) gcc options: -lrt
Phoronix Test Suite Self Test This is a wrapper around the phoronix-test-suite debug-self-test sub-command for testing hot Phoronix Test Suite code paths and if wanting to compare different PHP implementations. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Peak Memory Usage GRAPHITE LTO LTO GRAPHITE 110 220 330 440 550 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 500 500 500 500 500 500 500 500 Result Confidence
O2 O3 OpenBenchmarking.org MB, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Peak Memory Usage GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49 Min: 500.49 / Avg: 500.49 / Max: 500.49
clpeak Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Global Memory Bandwidth GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 145 145 145 145 145 145 145 145 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Global Memory Bandwidth GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 144.69 / Avg: 144.7 / Max: 144.7 Min: 144.57 / Avg: 144.64 / Max: 144.71 Min: 144.7 / Avg: 144.7 / Max: 144.7 Min: 144.67 / Avg: 144.69 / Max: 144.7 Min: 144.69 / Avg: 144.7 / Max: 144.71 Min: 144.69 / Avg: 144.7 / Max: 144.72 Min: 144.69 / Avg: 144.7 / Max: 144.7 Min: 144.69 / Avg: 144.7 / Max: 144.7 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Result
O2 O3 OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Double-Precision Double GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 144 144 144 144 144 144 144 144 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Double-Precision Double GRAPHITE LTO LTO GRAPHITE 30 60 90 120 150 Min: 144.07 / Avg: 144.19 / Max: 144.34 Min: 144.04 / Avg: 144.2 / Max: 144.39 Min: 144.18 / Avg: 144.3 / Max: 144.49 Min: 144.06 / Avg: 144.18 / Max: 144.33 Min: 144 / Avg: 144.16 / Max: 144.37 Min: 144.21 / Avg: 144.29 / Max: 144.43 Min: 144.07 / Avg: 144.21 / Max: 144.38 Min: 144.07 / Avg: 144.19 / Max: 144.35 1. (CXX) g++ options: -march=native -pipe -O3 -rdynamic -lOpenCL
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Decrypt GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 SE +/- 0.19, N = 3 SE +/- 0.06, N = 3 SE +/- 0.21, N = 3 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.35, N = 3 SE +/- 0.01, N = 3 295 295 295 295 295 295 295 295 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Decrypt GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 294.54 / Avg: 294.79 / Max: 295.16 Min: 294.56 / Avg: 294.64 / Max: 294.76 Min: 294.54 / Avg: 294.84 / Max: 295.25 Min: 294.56 / Avg: 294.71 / Max: 294.88 Min: 294.71 / Avg: 294.72 / Max: 294.74 Min: 294.55 / Avg: 294.7 / Max: 294.86 Min: 294.48 / Avg: 294.96 / Max: 295.65 Min: 294.59 / Avg: 294.61 / Max: 294.63 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Encrypt GRAPHITE LTO LTO GRAPHITE 60 120 180 240 300 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 296 296 296 296 296 296 296 296 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Encrypt GRAPHITE LTO LTO GRAPHITE 50 100 150 200 250 Min: 295.91 / Avg: 296.11 / Max: 296.22 Min: 295.91 / Avg: 295.99 / Max: 296.11 Min: 295.94 / Avg: 296.02 / Max: 296.15 Min: 295.92 / Avg: 296.03 / Max: 296.1 Min: 295.95 / Avg: 296.12 / Max: 296.23 Min: 296.02 / Avg: 296.1 / Max: 296.25 Min: 295.97 / Avg: 296.05 / Max: 296.12 Min: 295.92 / Avg: 295.97 / Max: 296.03 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Result
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Encrypt GRAPHITE LTO LTO GRAPHITE 90 180 270 360 450 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 401 401 401 401 401 401 401 401 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt Result Confidence
O2 O3 OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Encrypt GRAPHITE LTO LTO GRAPHITE 70 140 210 280 350 Min: 400.91 / Avg: 401.05 / Max: 401.13 Min: 401.09 / Avg: 401.28 / Max: 401.38 Min: 401.13 / Avg: 401.2 / Max: 401.26 Min: 401.24 / Avg: 401.3 / Max: 401.34 Min: 400.87 / Avg: 400.98 / Max: 401.07 Min: 400.82 / Avg: 401.02 / Max: 401.35 Min: 400.85 / Avg: 401.01 / Max: 401.13 Min: 401.42 / Avg: 401.45 / Max: 401.48 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Monte Carlo GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 SE +/- 0.08, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.14, N = 4 SE +/- 0.08, N = 4 SE +/- 0.07, N = 4 SE +/- 0.07, N = 4 SE +/- 0.26, N = 4 1085 1085 1085 1085 1085 1085 1085 1085 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Monte Carlo GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 1085.14 / Avg: 1085.27 / Max: 1085.41 Min: 1084.86 / Avg: 1084.86 / Max: 1084.86 Min: 1084.86 / Avg: 1084.86 / Max: 1084.86 Min: 1084.86 / Avg: 1085.27 / Max: 1085.41 Min: 1084.86 / Avg: 1085 / Max: 1085.14 Min: 1084.86 / Avg: 1085.07 / Max: 1085.14 Min: 1084.86 / Avg: 1084.93 / Max: 1085.14 Min: 1084.86 / Avg: 1085.2 / Max: 1085.96
BLAKE2 This is a benchmark of BLAKE2 using the blake2s binary. BLAKE2 is a high-performance crypto alternative to MD5 and SHA-2/3. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Cycles Per Byte, Fewer Is Better BLAKE2 20170307 GRAPHITE LTO LTO GRAPHITE 0.7605 1.521 2.2815 3.042 3.8025 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.38 3.38 3.38 3.38 3.38 3.38 3.38 3.38 1. (CC) gcc options: -O3 -march=native -lcrypto -lz Result Confidence
O2 O3 OpenBenchmarking.org Cycles Per Byte, Fewer Is Better BLAKE2 20170307 GRAPHITE LTO LTO GRAPHITE 2 4 6 8 10 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.38 Min: 3.38 / Avg: 3.38 / Max: 3.39 Min: 3.38 / Avg: 3.38 / Max: 3.38 1. (CC) gcc options: -O3 -march=native -lcrypto -lz
O2 O3 OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: XZ 0 - Process: Decompression GRAPHITE LTO LTO GRAPHITE 20 40 60 80 100 106 106 106 106 106 106 106 106 1. (CXX) g++ options: -lrt -static -lpthread
Schbench This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 424.17, N = 15 SE +/- 429.71, N = 12 SE +/- 320.32, N = 15 SE +/- 472.69, N = 12 SE +/- 547.15, N = 15 SE +/- 599.09, N = 15 SE +/- 577.60, N = 15 SE +/- 461.30, N = 15 18490 19509 19219 18633 17977 17286 18317 18057 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 4 - Workers Per Message Thread: 16 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 15728 / Avg: 18489.6 / Max: 21152 Min: 17504 / Avg: 19509.33 / Max: 21856 Min: 17248 / Avg: 19219.2 / Max: 22176 Min: 15568 / Avg: 18633.33 / Max: 21536 Min: 14416 / Avg: 17976.53 / Max: 21024 Min: 13008 / Avg: 17286.4 / Max: 20256 Min: 13008 / Avg: 18316.8 / Max: 21216 Min: 14288 / Avg: 18056.53 / Max: 20448 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 330.48, N = 15 SE +/- 377.81, N = 15 SE +/- 321.66, N = 15 SE +/- 391.99, N = 15 SE +/- 513.56, N = 12 SE +/- 398.91, N = 15 SE +/- 210.62, N = 13 SE +/- 410.48, N = 15 7907 7732 8213 7326 7216 7489 8178 7574 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 32 - Workers Per Message Thread: 2 GRAPHITE LTO LTO GRAPHITE 1400 2800 4200 5600 7000 Min: 6184 / Avg: 7907.2 / Max: 10416 Min: 4808 / Avg: 7732.27 / Max: 10000 Min: 5624 / Avg: 8213.33 / Max: 10864 Min: 4520 / Avg: 7325.87 / Max: 10512 Min: 4328 / Avg: 7216 / Max: 9904 Min: 4696 / Avg: 7488.53 / Max: 10832 Min: 6152 / Avg: 8177.85 / Max: 9104 Min: 3956 / Avg: 7573.6 / Max: 10000 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 216.51, N = 3 SE +/- 808.09, N = 15 SE +/- 1154.28, N = 12 SE +/- 1206.66, N = 12 SE +/- 760.86, N = 12 SE +/- 1012.51, N = 12 SE +/- 897.77, N = 15 SE +/- 1071.55, N = 15 25227 23290 23689 23991 24096 22020 23931 23092 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 2 - Workers Per Message Thread: 32 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 24800 / Avg: 25226.67 / Max: 25504 Min: 18720 / Avg: 23289.6 / Max: 27296 Min: 14096 / Avg: 23689.33 / Max: 28000 Min: 14256 / Avg: 23990.67 / Max: 28896 Min: 19552 / Avg: 24096 / Max: 28000 Min: 14992 / Avg: 22020 / Max: 28128 Min: 15664 / Avg: 23930.67 / Max: 27872 Min: 16208 / Avg: 23092.27 / Max: 28256 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 5K 10K 15K 20K 25K SE +/- 358.88, N = 3 SE +/- 423.93, N = 15 SE +/- 482.28, N = 15 SE +/- 442.32, N = 15 SE +/- 426.89, N = 15 SE +/- 354.90, N = 15 SE +/- 76.92, N = 3 SE +/- 305.69, N = 15 23157 23328 24036 23345 23196 23290 23221 22897 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K Min: 22624 / Avg: 23157.33 / Max: 23840 Min: 21984 / Avg: 23328 / Max: 28384 Min: 21984 / Avg: 24036.27 / Max: 28256 Min: 21600 / Avg: 23345.07 / Max: 27296 Min: 21024 / Avg: 23195.73 / Max: 27168 Min: 21152 / Avg: 23289.6 / Max: 25696 Min: 23072 / Avg: 23221.33 / Max: 23328 Min: 20896 / Avg: 22897.07 / Max: 24992 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 176.48, N = 4 SE +/- 138.67, N = 3 SE +/- 203.36, N = 15 SE +/- 129.39, N = 8 SE +/- 182.39, N = 15 SE +/- 265.87, N = 12 SE +/- 196.91, N = 15 SE +/- 186.50, N = 15 11232 10459 10798 10936 10794 10968 10881 10553 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 16 - Workers Per Message Thread: 4 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 10704 / Avg: 11232 / Max: 11440 Min: 10224 / Avg: 10458.67 / Max: 10704 Min: 9072 / Avg: 10797.87 / Max: 11984 Min: 10384 / Avg: 10936 / Max: 11472 Min: 9520 / Avg: 10793.6 / Max: 11888 Min: 9904 / Avg: 10968 / Max: 13328 Min: 9232 / Avg: 10881.07 / Max: 11888 Min: 9008 / Avg: 10552.53 / Max: 11664 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 488.50, N = 15 SE +/- 267.24, N = 15 SE +/- 201.78, N = 5 SE +/- 453.98, N = 15 SE +/- 21.33, N = 3 SE +/- 560.84, N = 15 SE +/- 241.52, N = 15 SE +/- 334.44, N = 15 14314 13494 13219 14336 12987 14683 13479 14165 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 8 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 12656 / Avg: 14313.6 / Max: 19232 Min: 12688 / Avg: 13494.4 / Max: 16112 Min: 12880 / Avg: 13219.2 / Max: 14000 Min: 12528 / Avg: 14336 / Max: 18016 Min: 12944 / Avg: 12986.67 / Max: 13008 Min: 12784 / Avg: 14682.67 / Max: 18784 Min: 12688 / Avg: 13479.47 / Max: 15472 Min: 12912 / Avg: 14165.33 / Max: 17184 1. (CC) gcc options: -O2 -lpthread
Result
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 211.49, N = 15 SE +/- 239.88, N = 15 SE +/- 207.78, N = 15 SE +/- 132.25, N = 15 SE +/- 233.63, N = 15 SE +/- 145.80, N = 15 SE +/- 118.00, N = 11 SE +/- 211.54, N = 15 10962 11193 10988 11118 11075 11067 11478 10757 1. (CC) gcc options: -O2 -lpthread Result Confidence
O2 O3 OpenBenchmarking.org usec, 99.9th Latency Percentile, Fewer Is Better Schbench Message Threads: 8 - Workers Per Message Thread: 6 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 8944 / Avg: 10962.13 / Max: 11920 Min: 8720 / Avg: 11192.53 / Max: 12016 Min: 9616 / Avg: 10987.73 / Max: 12240 Min: 10288 / Avg: 11117.87 / Max: 12112 Min: 9712 / Avg: 11075.2 / Max: 12016 Min: 10064 / Avg: 11066.67 / Max: 11760 Min: 10896 / Avg: 11477.82 / Max: 12080 Min: 9616 / Avg: 10757.33 / Max: 12080 1. (CC) gcc options: -O2 -lpthread
RAR Compression This test measures the time needed to archive/compress two copies of the Linux 4.13 kernel source tree using RAR/WinRAR compression. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better RAR Compression 5.6.1 Linux Source Tree Archiving To RAR GRAPHITE LTO LTO GRAPHITE 15 30 45 60 75 SE +/- 1.83, N = 12 SE +/- 1.83, N = 12 SE +/- 1.84, N = 12 SE +/- 1.86, N = 12 SE +/- 1.86, N = 12 SE +/- 1.86, N = 12 SE +/- 1.85, N = 12 SE +/- 1.83, N = 12 68.18 68.22 68.06 68.12 68.06 68.20 68.11 68.14 Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better RAR Compression 5.6.1 Linux Source Tree Archiving To RAR GRAPHITE LTO LTO GRAPHITE 13 26 39 52 65 Min: 48.09 / Avg: 68.18 / Max: 70.25 Min: 48.12 / Avg: 68.22 / Max: 70.2 Min: 47.86 / Avg: 68.06 / Max: 70.1 Min: 47.68 / Avg: 68.12 / Max: 70.25 Min: 47.58 / Avg: 68.06 / Max: 70.29 Min: 47.75 / Avg: 68.2 / Max: 70.35 Min: 47.8 / Avg: 68.11 / Max: 70.17 Min: 48.06 / Avg: 68.14 / Max: 70.13
Redis Redis is an open-source data structure server. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K SE +/- 49354.23, N = 12 SE +/- 22278.31, N = 3 SE +/- 42636.67, N = 14 SE +/- 41556.34, N = 15 SE +/- 44000.06, N = 15 SE +/- 12198.66, N = 3 SE +/- 36074.39, N = 15 SE +/- 42914.21, N = 15 1656333 1770465 1637953 1670885 1637539 1873988 1605114 1740022 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K Min: 1422475.12 / Avg: 1656333.06 / Max: 1862197.38 Min: 1745200.62 / Avg: 1770465.37 / Max: 1814882 Min: 1445086.75 / Avg: 1637952.7 / Max: 1897533.12 Min: 1443001.38 / Avg: 1670885.21 / Max: 1841620.62 Min: 1468428.75 / Avg: 1637538.96 / Max: 1872659.25 Min: 1851851.75 / Avg: 1873987.92 / Max: 1893939.38 Min: 1447178 / Avg: 1605113.8 / Max: 1814882 Min: 1485884.12 / Avg: 1740021.65 / Max: 1919385.75 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET GRAPHITE LTO LTO GRAPHITE 500K 1000K 1500K 2000K 2500K SE +/- 35503.43, N = 15 SE +/- 32280.60, N = 7 SE +/- 39645.32, N = 4 SE +/- 47452.96, N = 15 SE +/- 36001.71, N = 5 SE +/- 46039.99, N = 15 SE +/- 37031.01, N = 15 SE +/- 46590.30, N = 15 2458302 2495376 2413016 2383136 2371908 2470630 2315232 2337500 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K Min: 2123142.25 / Avg: 2458301.57 / Max: 2638522.25 Min: 2369668.25 / Avg: 2495376.46 / Max: 2638522.25 Min: 2341920.5 / Avg: 2413015.69 / Max: 2518891.5 Min: 1915708.75 / Avg: 2383135.87 / Max: 2624671.75 Min: 2242152.5 / Avg: 2371907.55 / Max: 2463054.25 Min: 2061855.62 / Avg: 2470629.57 / Max: 2673796.75 Min: 2070393.38 / Avg: 2315231.72 / Max: 2500000 Min: 1976284.62 / Avg: 2337500.45 / Max: 2564102.75 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K SE +/- 27901.87, N = 12 SE +/- 17642.80, N = 15 SE +/- 36092.91, N = 15 SE +/- 36228.58, N = 12 SE +/- 32251.87, N = 15 SE +/- 26391.54, N = 15 SE +/- 3926.71, N = 3 SE +/- 10444.11, N = 3 1760281 1750626 1565614 1620389 1620645 1543733 1532979 1422628 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K Min: 1468428.75 / Avg: 1760281.09 / Max: 1808318.38 Min: 1615508.88 / Avg: 1750626.32 / Max: 1805054.12 Min: 1424501.38 / Avg: 1565613.81 / Max: 1824817.62 Min: 1420454.5 / Avg: 1620388.69 / Max: 1776198.88 Min: 1432664.75 / Avg: 1620644.82 / Max: 1814882 Min: 1414427.12 / Avg: 1543733.23 / Max: 1798561.12 Min: 1529052 / Avg: 1532978.71 / Max: 1540832.12 Min: 1408450.75 / Avg: 1422627.58 / Max: 1443001.38 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD GRAPHITE LTO LTO GRAPHITE 500K 1000K 1500K 2000K 2500K SE +/- 58274.33, N = 15 SE +/- 53674.02, N = 15 SE +/- 16766.14, N = 3 SE +/- 52644.10, N = 15 SE +/- 51368.48, N = 15 SE +/- 6440.58, N = 3 SE +/- 44604.29, N = 15 SE +/- 38089.22, N = 3 1935818 1923067 2193241 1927131 1916520 2194624 1915884 2043649 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K Min: 1663893.5 / Avg: 1935817.98 / Max: 2217295 Min: 1639344.25 / Avg: 1923067.08 / Max: 2227171.5 Min: 2159827.25 / Avg: 2193240.75 / Max: 2212389.5 Min: 1636661.25 / Avg: 1927130.71 / Max: 2222222.25 Min: 1663893.5 / Avg: 1916520.2 / Max: 2183406 Min: 2188183.75 / Avg: 2194624.33 / Max: 2207505.5 Min: 1669449.12 / Avg: 1915883.97 / Max: 2118644 Min: 1968503.88 / Avg: 2043649.17 / Max: 2092050.25 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Result
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP GRAPHITE LTO LTO GRAPHITE 600K 1200K 1800K 2400K 3000K SE +/- 50781.01, N = 15 SE +/- 50196.02, N = 15 SE +/- 58370.54, N = 15 SE +/- 32312.34, N = 15 SE +/- 45866.07, N = 3 SE +/- 13774.42, N = 3 SE +/- 37834.45, N = 15 SE +/- 51584.55, N = 3 2313264 2400678 2394427 2476071 2446730 2790833 2408656 2559594 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread Result Confidence
O2 O3 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP GRAPHITE LTO LTO GRAPHITE 500K 1000K 1500K 2000K 2500K Min: 1968503.88 / Avg: 2313264.34 / Max: 2624671.75 Min: 2008032.12 / Avg: 2400678.09 / Max: 2597402.75 Min: 2044989.75 / Avg: 2394427.38 / Max: 2695417.75 Min: 2202643.25 / Avg: 2476070.62 / Max: 2624671.75 Min: 2358490.5 / Avg: 2446729.67 / Max: 2512562.75 Min: 2770083 / Avg: 2790833.25 / Max: 2816901.5 Min: 2114165 / Avg: 2408655.65 / Max: 2631579 Min: 2487562.25 / Avg: 2559594.08 / Max: 2659574.5 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 300 600 900 1200 1500 SE +/- 131.09, N = 4 SE +/- 139.03, N = 4 SE +/- 129.30, N = 4 SE +/- 2.54, N = 4 SE +/- 158.10, N = 4 SE +/- 149.55, N = 4 SE +/- 128.60, N = 4 SE +/- 130.31, N = 4 1207 1223 1203 810 1082 1074 943 942 Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Fast Fourier Transform GRAPHITE LTO LTO GRAPHITE 200 400 600 800 1000 Min: 814.69 / Avg: 1206.95 / Max: 1363.75 Min: 809.02 / Avg: 1222.58 / Max: 1411.2 Min: 814.69 / Avg: 1202.55 / Max: 1337.61 Min: 804.54 / Avg: 810.07 / Max: 815.45 Min: 807.89 / Avg: 1082.3 / Max: 1370.18 Min: 814.3 / Avg: 1073.5 / Max: 1337.1 Min: 813.17 / Avg: 942.65 / Max: 1328.44 Min: 808.27 / Avg: 942.09 / Max: 1333.01
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K SE +/- 570.36, N = 15 SE +/- 536.99, N = 15 SE +/- 659.06, N = 15 SE +/- 686.03, N = 3 SE +/- 584.22, N = 15 SE +/- 369.66, N = 3 SE +/- 67.22, N = 3 SE +/- 860.85, N = 15 35459 34465 34599 35309 34406 33337 33711 35307 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 256 GRAPHITE LTO LTO GRAPHITE 6K 12K 18K 24K 30K Min: 32314 / Avg: 35458.87 / Max: 39471 Min: 30091 / Avg: 34464.67 / Max: 37872 Min: 28792 / Avg: 34598.93 / Max: 39307 Min: 34466 / Avg: 35309 / Max: 36668 Min: 30175 / Avg: 34405.67 / Max: 38624 Min: 32675 / Avg: 33337.33 / Max: 33953 Min: 33583 / Avg: 33710.67 / Max: 33811 Min: 29720 / Avg: 35306.87 / Max: 39849 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 797.17, N = 15 SE +/- 855.41, N = 15 SE +/- 676.98, N = 15 SE +/- 742.73, N = 15 SE +/- 874.72, N = 15 SE +/- 25.10, N = 3 SE +/- 570.27, N = 12 SE +/- 587.53, N = 15 36030 37112 38696 37417 37135 39689 39679 37825 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 7K 14K 21K 28K 35K Min: 30253 / Avg: 36029.67 / Max: 39581 Min: 32524 / Avg: 37112.2 / Max: 40512 Min: 33419 / Avg: 38696.33 / Max: 40450 Min: 33762 / Avg: 37417.07 / Max: 40385 Min: 32734 / Avg: 37135.13 / Max: 40482 Min: 39656 / Avg: 39688.67 / Max: 39738 Min: 33622 / Avg: 39678.67 / Max: 41235 Min: 33537 / Avg: 37825.13 / Max: 40450 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K SE +/- 718.14, N = 15 SE +/- 771.02, N = 15 SE +/- 629.27, N = 15 SE +/- 241.99, N = 3 SE +/- 619.65, N = 15 SE +/- 615.72, N = 15 SE +/- 693.88, N = 15 SE +/- 430.35, N = 3 40987 40327 39837 38412 40094 40046 40773 43982 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 8K 16K 24K 32K 40K Min: 38568 / Avg: 40987.13 / Max: 45361 Min: 36860 / Avg: 40327.47 / Max: 45239 Min: 36815 / Avg: 39837.2 / Max: 44578 Min: 38106 / Avg: 38412.33 / Max: 38890 Min: 37739 / Avg: 40093.6 / Max: 44630 Min: 37719 / Avg: 40045.53 / Max: 43684 Min: 36944 / Avg: 40772.53 / Max: 44178 Min: 43530 / Avg: 43981.67 / Max: 44842 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 4K 8K 12K 16K 20K SE +/- 307.79, N = 4 SE +/- 450.15, N = 15 SE +/- 505.45, N = 15 SE +/- 403.51, N = 15 SE +/- 610.20, N = 12 SE +/- 484.52, N = 15 SE +/- 604.04, N = 15 SE +/- 322.28, N = 3 18815 19061 19269 18955 18880 19063 19224 18619 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K Min: 18243 / Avg: 18815 / Max: 19687 Min: 16534 / Avg: 19061 / Max: 22643 Min: 16944 / Avg: 19268.6 / Max: 22995 Min: 16444 / Avg: 18955.33 / Max: 23373 Min: 15821 / Avg: 18879.75 / Max: 22588 Min: 15913 / Avg: 19062.67 / Max: 22920 Min: 16377 / Avg: 19223.93 / Max: 23107 Min: 17991 / Avg: 18619.33 / Max: 19058 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 253.60, N = 15 SE +/- 90.42, N = 3 SE +/- 270.70, N = 15 SE +/- 244.22, N = 15 SE +/- 260.33, N = 15 SE +/- 166.28, N = 3 SE +/- 195.10, N = 15 SE +/- 242.50, N = 15 10170 10642 10225 10418 9866 10070 10584 10672 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 4096 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 8247.3 / Avg: 10169.98 / Max: 10865 Min: 10482 / Avg: 10642 / Max: 10795 Min: 9068.8 / Avg: 10224.55 / Max: 11777 Min: 9181.9 / Avg: 10417.67 / Max: 11679 Min: 8188.8 / Avg: 9865.81 / Max: 10685 Min: 9882.1 / Avg: 10070.47 / Max: 10402 Min: 8920.9 / Avg: 10584.45 / Max: 11434 Min: 8856.8 / Avg: 10671.83 / Max: 11630 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 203.40, N = 15 SE +/- 278.07, N = 15 SE +/- 245.30, N = 15 SE +/- 281.26, N = 15 SE +/- 14.47, N = 3 SE +/- 235.96, N = 15 SE +/- 162.08, N = 3 SE +/- 390.04, N = 12 9960 9486 10093 9881 10837 10208 11284 9326 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 2048 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 8197.2 / Avg: 9960.23 / Max: 10895 Min: 7479.6 / Avg: 9486.15 / Max: 10744 Min: 9148.4 / Avg: 10092.63 / Max: 11803 Min: 8017.2 / Avg: 9881.02 / Max: 11494 Min: 10813 / Avg: 10837 / Max: 10863 Min: 8030.6 / Avg: 10207.79 / Max: 10845 Min: 10960 / Avg: 11283.67 / Max: 11461 Min: 7491.8 / Avg: 9325.83 / Max: 11313 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 6.51, N = 3 SE +/- 15.86, N = 3 SE +/- 262.72, N = 15 SE +/- 37.81, N = 3 SE +/- 169.10, N = 15 SE +/- 19.65, N = 3 SE +/- 225.24, N = 15 SE +/- 78.04, N = 3 10404 10413 10526 11200 10142 10432 10588 11068 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 10397 / Avg: 10404 / Max: 10417 Min: 10396 / Avg: 10413.33 / Max: 10445 Min: 8842 / Avg: 10526.13 / Max: 11348 Min: 11147 / Avg: 11199.67 / Max: 11273 Min: 8587.1 / Avg: 10142.39 / Max: 10557 Min: 10394 / Avg: 10432.33 / Max: 10459 Min: 9171.5 / Avg: 10587.51 / Max: 11307 Min: 10937 / Avg: 11068 / Max: 11207 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K SE +/- 14.69, N = 3 SE +/- 251.97, N = 15 SE +/- 231.71, N = 15 SE +/- 217.22, N = 15 SE +/- 173.76, N = 3 SE +/- 124.91, N = 5 SE +/- 212.32, N = 12 SE +/- 51.23, N = 3 9291 9646 9950 9889 8918 9057 9817 9715 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 9275.3 / Avg: 9291.37 / Max: 9320.7 Min: 9093 / Avg: 9645.51 / Max: 11945 Min: 9214.9 / Avg: 9950.09 / Max: 12194 Min: 9466.5 / Avg: 9888.65 / Max: 12024 Min: 8570.3 / Avg: 8917.67 / Max: 9100.2 Min: 8569.1 / Avg: 9057.38 / Max: 9252.8 Min: 9208.8 / Avg: 9816.68 / Max: 12105 Min: 9638.4 / Avg: 9714.57 / Max: 9812 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 17.85, N = 3 SE +/- 436.96, N = 15 SE +/- 21.50, N = 3 SE +/- 81.26, N = 3 SE +/- 613.50, N = 15 SE +/- 621.38, N = 15 SE +/- 28.84, N = 3 SE +/- 43.26, N = 3 11267 11731 11248 11134 13660 13622 11112 11015 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 32 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 11248 / Avg: 11267.33 / Max: 11303 Min: 10005 / Avg: 11730.53 / Max: 15840 Min: 11220 / Avg: 11247.67 / Max: 11290 Min: 10973 / Avg: 11134.33 / Max: 11232 Min: 11131 / Avg: 13660.2 / Max: 15907 Min: 10819 / Avg: 13622.33 / Max: 15892 Min: 11056 / Avg: 11112 / Max: 11152 Min: 10962 / Avg: 11015.33 / Max: 11101 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Result
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 3K 6K 9K 12K 15K SE +/- 32.87, N = 3 SE +/- 630.65, N = 15 SE +/- 195.21, N = 15 SE +/- 217.77, N = 15 SE +/- 510.16, N = 15 SE +/- 8.74, N = 3 SE +/- 300.19, N = 15 SE +/- 227.38, N = 15 13780 11990 10409 10175 12880 14087 10207 10232 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm Result Confidence
O2 O3 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 64 GRAPHITE LTO LTO GRAPHITE 2K 4K 6K 8K 10K Min: 13714 / Avg: 13779.67 / Max: 13815 Min: 9000 / Avg: 11990.06 / Max: 14067 Min: 9253.5 / Avg: 10409.47 / Max: 11419 Min: 8487.1 / Avg: 10174.88 / Max: 11209 Min: 9033 / Avg: 12879.52 / Max: 13935 Min: 14076 / Avg: 14086.67 / Max: 14104 Min: 9041.4 / Avg: 10207.23 / Max: 11544 Min: 8969.9 / Avg: 10231.65 / Max: 11187 1. (CC) gcc options: -pthread -march=native -pipe -O3 -lm
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 SE +/- 0.19, N = 15 SE +/- 0.01, N = 3 SE +/- 0.20, N = 15 SE +/- 0.18, N = 15 SE +/- 0.18, N = 15 SE +/- 0.20, N = 15 SE +/- 0.01, N = 3 SE +/- 0.16, N = 15 11.84 10.83 11.40 11.21 11.25 11.70 10.80 11.09 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
O2 O3 OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster GRAPHITE LTO LTO GRAPHITE 3 6 9 12 15 Min: 10.83 / Avg: 11.84 / Max: 12.37 Min: 10.81 / Avg: 10.83 / Max: 10.85 Min: 10.77 / Avg: 11.4 / Max: 12.33 Min: 10.79 / Avg: 11.21 / Max: 12.32 Min: 10.82 / Avg: 11.25 / Max: 12.39 Min: 10.8 / Avg: 11.7 / Max: 12.33 Min: 10.79 / Avg: 10.8 / Max: 10.81 Min: 10.76 / Avg: 11.09 / Max: 12.28 1. (CXX) g++ options: -O2 -lOpenCL
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: EP.C GRAPHITE LTO LTO GRAPHITE 600 1200 1800 2400 3000 SE +/- 8.95, N = 3 SE +/- 11.67, N = 3 SE +/- 41.25, N = 15 SE +/- 41.34, N = 15 SE +/- 16.80, N = 3 SE +/- 35.23, N = 3 SE +/- 46.16, N = 15 SE +/- 65.01, N = 15 2715 2695 2496 2449 2709 2682 2497 2515 1. 3.3 Result Confidence
O2 O3 OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: EP.C GRAPHITE LTO LTO GRAPHITE 500 1000 1500 2000 2500 Min: 2701.02 / Avg: 2715.39 / Max: 2731.82 Min: 2679.79 / Avg: 2695.4 / Max: 2718.23 Min: 2327.18 / Avg: 2495.59 / Max: 2714.2 Min: 2341.4 / Avg: 2449.04 / Max: 2747.29 Min: 2685.6 / Avg: 2708.74 / Max: 2741.41 Min: 2617.9 / Avg: 2681.56 / Max: 2739.53 Min: 2310.2 / Avg: 2496.68 / Max: 2711.92 Min: 1821.16 / Avg: 2515.18 / Max: 2713.57 1. 3.3
Threaded I/O Tester Tiotester (Threaded I/O Tester) benchmarks the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 600K 1200K 1800K 2400K 3000K SE +/- 37971.60, N = 3 SE +/- 32461.38, N = 3 SE +/- 39625.19, N = 15 SE +/- 41740.39, N = 15 SE +/- 24115.47, N = 3 SE +/- 17175.69, N = 3 SE +/- 59239.72, N = 12 SE +/- 53296.44, N = 15 2535256 2530270 2453843 2481843 2493472 2593178 2357395 2322361 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 400K 800K 1200K 1600K 2000K Min: 2482424.24 / Avg: 2535255.82 / Max: 2608917.2 Min: 2473429.95 / Avg: 2530269.64 / Max: 2585858.59 Min: 2211663.07 / Avg: 2453843.39 / Max: 2661468.49 Min: 2193893.95 / Avg: 2481843.24 / Max: 2769438.81 Min: 2445373.13 / Avg: 2493471.52 / Max: 2520615.39 Min: 2572864.32 / Avg: 2593178.21 / Max: 2627325.21 Min: 1995129.08 / Avg: 2357395.05 / Max: 2668403.91 Min: 2014756.52 / Avg: 2322360.85 / Max: 2656290.53 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 300K 600K 900K 1200K 1500K SE +/- 21930.96, N = 4 SE +/- 14690.71, N = 3 SE +/- 26025.10, N = 15 SE +/- 22684.71, N = 15 SE +/- 26485.39, N = 15 SE +/- 36089.61, N = 15 SE +/- 30751.89, N = 15 SE +/- 17982.37, N = 15 1299985 1294356 1241428 1230864 1250589 1191378 1211522 1258276 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 128MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 200K 400K 600K 800K 1000K Min: 1254901.96 / Avg: 1299984.92 / Max: 1343832.02 Min: 1264978.38 / Avg: 1294355.83 / Max: 1309462.92 Min: 986987.95 / Avg: 1241428.17 / Max: 1356291.39 Min: 1110629.07 / Avg: 1230863.93 / Max: 1332465.84 Min: 1013359.72 / Avg: 1250588.73 / Max: 1360797.34 Min: 831844.03 / Avg: 1191378.08 / Max: 1343832.02 Min: 951230.84 / Avg: 1211521.93 / Max: 1417301.04 Min: 1131491.71 / Avg: 1258275.81 / Max: 1329870.13 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 140K 280K 420K 560K 700K SE +/- 11328.85, N = 15 SE +/- 10224.24, N = 15 SE +/- 12516.07, N = 15 SE +/- 11524.05, N = 15 SE +/- 12258.57, N = 15 SE +/- 11753.75, N = 15 SE +/- 12846.91, N = 15 SE +/- 16476.15, N = 15 631416 644903 639169 638222 636360 651278 635045 623268 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 110K 220K 330K 440K 550K Min: 561403.51 / Avg: 631415.97 / Max: 685408.3 Min: 565121.41 / Avg: 644902.66 / Max: 690958.17 Min: 571428.57 / Avg: 639169.27 / Max: 712595.69 Min: 564187.33 / Avg: 638222.47 / Max: 690492.25 Min: 544680.85 / Avg: 636359.88 / Max: 690958.17 Min: 578858.11 / Avg: 651278.44 / Max: 709141.27 Min: 557430.59 / Avg: 635045.14 / Max: 711605.28 Min: 522715.67 / Avg: 623268.06 / Max: 703780.07 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 70K 140K 210K 280K 350K SE +/- 5093.99, N = 15 SE +/- 12738.48, N = 12 SE +/- 6744.06, N = 15 SE +/- 5089.98, N = 15 SE +/- 6970.34, N = 15 SE +/- 7607.10, N = 15 SE +/- 7551.52, N = 15 SE +/- 7389.89, N = 15 341106 339063 320411 334458 328113 336770 331940 315850 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 60K 120K 180K 240K 300K Min: 297328.69 / Avg: 341105.69 / Max: 365975.7 Min: 212801.33 / Avg: 339062.71 / Max: 367288.38 Min: 284760.85 / Avg: 320411.33 / Max: 362863.22 Min: 299590.4 / Avg: 334457.98 / Max: 363378.28 Min: 255616.58 / Avg: 328112.55 / Max: 353103.45 Min: 260559.8 / Avg: 336769.55 / Max: 365453.25 Min: 256256.26 / Avg: 331939.81 / Max: 360563.38 Min: 281473.34 / Avg: 315849.96 / Max: 361071.93 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 80K 160K 240K 320K 400K SE +/- 7177.12, N = 3 SE +/- 8294.11, N = 15 SE +/- 6150.11, N = 15 SE +/- 3838.15, N = 11 SE +/- 8701.27, N = 15 SE +/- 5195.20, N = 3 SE +/- 5494.13, N = 15 SE +/- 6425.45, N = 15 390904 380026 379322 379349 365436 387137 378003 371220 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Random Read - Size Per Thread: 64MB - Thread Count: 8 GRAPHITE LTO LTO GRAPHITE 70K 140K 210K 280K 350K Min: 378978.53 / Avg: 390903.57 / Max: 403785.49 Min: 307322.93 / Avg: 380025.57 / Max: 415247.36 Min: 321406.15 / Avg: 379321.94 / Max: 412238.33 Min: 342704.15 / Avg: 379348.61 / Max: 388467.38 Min: 315270.94 / Avg: 365435.55 / Max: 410585.41 Min: 381520.12 / Avg: 387136.85 / Max: 397515.53 Min: 326738.99 / Avg: 378002.98 / Max: 411244.98 Min: 321608.04 / Avg: 371219.5 / Max: 400312.74 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 13K 26K 39K 52K 65K SE +/- 1376.66, N = 15 SE +/- 1666.45, N = 15 SE +/- 1570.27, N = 15 SE +/- 113.41, N = 3 SE +/- 98.77, N = 3 SE +/- 1850.70, N = 15 SE +/- 1574.97, N = 15 SE +/- 1278.46, N = 15 57032 55352 55936 59090 59348 54237 55920 57105 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 256MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 43836.56 / Avg: 57032.1 / Max: 59300.44 Min: 44570.67 / Avg: 55351.67 / Max: 59316.76 Min: 43804.21 / Avg: 55935.61 / Max: 59221.56 Min: 58932.71 / Avg: 59089.57 / Max: 59309.89 Min: 59150.58 / Avg: 59348.03 / Max: 59451.93 Min: 44033.07 / Avg: 54237.18 / Max: 59341.68 Min: 44033.54 / Avg: 55919.5 / Max: 59185.62 Min: 44908.34 / Avg: 57105.34 / Max: 59261.83 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 1580.42, N = 15 SE +/- 1574.63, N = 15 SE +/- 1623.64, N = 15 SE +/- 1124.24, N = 12 SE +/- 83.63, N = 3 SE +/- 154.78, N = 3 SE +/- 234.51, N = 3 SE +/- 1453.76, N = 15 52890 52643 52892 55233 56985 56619 56108 53677 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 64MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 10K 20K 30K 40K 50K Min: 42431.53 / Avg: 52889.81 / Max: 56911.02 Min: 42115.65 / Avg: 52643.1 / Max: 56637.17 Min: 42585.05 / Avg: 52892.23 / Max: 56869.93 Min: 42969.24 / Avg: 55233.48 / Max: 57028.29 Min: 56901.53 / Avg: 56985.16 / Max: 57152.43 Min: 56325.63 / Avg: 56619.23 / Max: 56850.99 Min: 55724.86 / Avg: 56108.47 / Max: 56533.98 Min: 42537.28 / Avg: 53676.84 / Max: 56806.84 1. (CC) gcc options: -O2
Result
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 12K 24K 36K 48K 60K SE +/- 1413.44, N = 15 SE +/- 263.41, N = 3 SE +/- 1561.99, N = 15 SE +/- 438.11, N = 3 SE +/- 1627.34, N = 15 SE +/- 1068.14, N = 12 SE +/- 1441.47, N = 15 SE +/- 1607.65, N = 15 51885 53941 51338 54386 50689 53618 51224 51196 1. (CC) gcc options: -O2 Result Confidence
O2 O3 OpenBenchmarking.org MB/s, More Is Better Threaded I/O Tester 20170503 Test: Read - Size Per Thread: 32MB - Thread Count: 16 GRAPHITE LTO LTO GRAPHITE 9K 18K 27K 36K 45K Min: 41160.87 / Avg: 51884.54 / Max: 55065.61 Min: 53522.89 / Avg: 53940.83 / Max: 54427.55 Min: 41313.65 / Avg: 51338.37 / Max: 55184.31 Min: 53590.12 / Avg: 54386.46 / Max: 55101.16 Min: 41602.34 / Avg: 50689.41 / Max: 55519.41 Min: 42022.32 / Avg: 53617.73 / Max: 55417.25 Min: 41497.81 / Avg: 51223.95 / Max: 55231.93 Min: 40461.51 / Avg: 51196 / Max: 55501.36 1. (CC) gcc options: -O2
O3: LTO GRAPHITE Environment Notes: FFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize CXXFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize CFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimizeCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 20 August 2019 09:39 by user wolfwood.
O3: LTO Environment Notes: FFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin CXXFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-plugin CFLAGS=-march=native-pipe-O3-flto=36-fuse-linker-pluginCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 21 August 2019 20:11 by user wolfwood.
O3: GRAPHITE Environment Notes: FFLAGS=-march=native-pipe-O3-fgraphite-identity-floop-nest-optimize CXXFLAGS=-march=native-pipe-O3-fgraphite-identity-floop-nest-optimize LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O3-fgraphite-identity-floop-nest-optimize CFLAGS=-march=native-pipe-O3-fgraphite-identity-floop-nest-optimizeCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 23 August 2019 21:56 by user wolfwood.
O3: Environment Notes: FFLAGS=-march=native-pipe-O3 CXXFLAGS=-march=native-pipe-O3 LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O3 CFLAGS=-march=native-pipe-O3Compiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 25 August 2019 09:04 by user wolfwood.
O2: LTO GRAPHITE Environment Notes: FFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize CXXFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimize CFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin-fgraphite-identity-floop-nest-optimizeCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 26 August 2019 21:11 by user wolfwood.
O2: LTO Environment Notes: FFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin CXXFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-plugin CFLAGS=-march=native-pipe-O2-ftree-vectorize-flto=36-fuse-linker-pluginCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 1 September 2019 17:45 by user wolfwood.
O2: GRAPHITE Environment Notes: FFLAGS=-march=native-pipe-O2-ftree-vectorize-fgraphite-identity-floop-nest-optimize CXXFLAGS=-march=native-pipe-O2-ftree-vectorize-fgraphite-identity-floop-nest-optimize LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O2-ftree-vectorize-fgraphite-identity-floop-nest-optimize CFLAGS=-march=native-pipe-O2-ftree-vectorize-fgraphite-identity-floop-nest-optimizeCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 2 September 2019 23:38 by user wolfwood.
O2: Processor: Intel Core i9-7980XE @ 4.40GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Taichi XE (P1.60 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 129024MB, Disk: 960GB INTEL SSDPED1D960GAY + 2000GB Western Digital WD20EARS-00S + 2000GB Seagate ST2000DM001-1CH1 + 3 x 2000GB Seagate ST2000DM001-1ER1 + 2000GB Western Digital WD20EARS-00M + 250GB HITACHI HTS54322, Graphics: eVGA NVIDIA GeForce GTX 970 4GB, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211 + Intel Dual Band-AC 3168NGW
OS: Gentoo/Linux, Kernel: 5.0.17-gentoo (x86_64), Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, Vulkan: 1.1.99, Compiler: GCC 9.2.0 + Clang 8.0.1 + LLVM 8.0.1 + CUDA 10.1, File-System: ext4, Screen Resolution: 1920x1080
Environment Notes: FFLAGS=-march=native-pipe-O2-ftree-vectorize CXXFLAGS=-march=native-pipe-O2-ftree-vectorize LDFLAGS=-Wl-O1-Wl--as-needed FCFLAGS=-march=native-pipe-O2-ftree-vectorize CFLAGS=-march=native-pipe-O2-ftree-vectorizeCompiler Notes: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/9.2.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0 --disable-altivec --disable-esp --disable-fixed-point --disable-isl-version-check --disable-libmudflap --disable-libssp --disable-systemtap --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-languages=c,c++,d,fortran --enable-libgomp --enable-libstdcxx-time --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/man --with-build-config=bootstrap-lto --with-isl --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/9.2.0/pythonDisk Notes: NONE / noatime,rwProcessor Notes: Scaling Governor: intel_pstate performanceJava Notes: OpenJDK Runtime Environment (IcedTea 3.12.0) (Gentoo icedtea-3.12.0)Python Notes: Python 3.6.9Security Notes: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling
Testing initiated at 29 August 2019 18:26 by user wolfwood.