AMD EPYC compiler benchmarks.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1807210-RA-AMDEPYCCO34 AMD EPYC Compiler Testing July 2018 - Phoronix Test Suite AMD EPYC Compiler Testing July 2018 AMD EPYC compiler benchmarks.
HTML result view exported from: https://openbenchmarking.org/result/1807210-RA-AMDEPYCCO34&export=pdf&grr&rdt&rro .
AMD EPYC Compiler Testing July 2018 Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Compiler File-System Screen Resolution GCC 8.2 RC1: -O3 -march=native AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native - PGO GCC 8.2 RC1: -O3 GCC 9.0 SVN: -O3 -march=native GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads) TYAN B8026T70AE24HR (V1.02.B10 BIOS) AMD Family 17h 129024MB 280GB INTEL SSDPE21D280GA ASPEED ASPEED Family VE228 Broadcom Limited NetXtreme BCM5720 Gigabit PCIe Ubuntu 18.10 4.16.0-041600-generic (x86_64) GCC 8.1.1 20180720 ext4 1920x1080 GCC 9.0.0 20180720 OpenBenchmarking.org Environment Details - GCC 8.2 RC1: -O3 -march=native: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=native - AMD EPYC 7601 32-Core: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=native - GCC 8.2 RC1: -O3 -march=native - PGO: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=native - GCC 8.2 RC1: -O3: CXXFLAGS=-O3 CFLAGS=-O3 - GCC 9.0 SVN: -O3 -march=native: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=native - GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=native Compiler Details - --disable-multilib --enable-checking=release Disk Details - NONE / data=ordered,errors=remount-ro,relatime,rw Processor Details - Scaling Governor: acpi-cpufreq ondemand Security Details - __user pointer sanitization + Full AMD retpoline IBPB Protection
AMD EPYC Compiler Testing July 2018 vpxenc: vpxenc m-queens: Time To Solve stockfish: Total Time ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 pgbench: Buffer Test - Single Thread - Read Write pgbench: Buffer Test - Normal Load - Read Only pgbench: Buffer Test - Single Thread - Read Only pgbench: Buffer Test - Normal Load - Read Write smallpt: Global Illumination Renderer; 100 Samples c-ray: Total Time tinymembench: Standard Memcpy compress-7zip: Compress Speed Test himeno: Poisson Pressure Solver apache: Static Web Page Serving graphics-magick: Sharpen graphics-magick: Local Adaptive Thresholding graphics-magick: Resizing graphics-magick: Blur graphics-magick: HWB Color Space aobench: 2048 x 2048 - Total Time hmmer: Pfam Database Search ebizzy: crafty: Elapsed Time scimark2: Composite redis: GET redis: SET sockperf: Latency Under Load openssl: RSA 4096-bit Performance redis: LPUSH redis: SADD encode-flac: WAV To FLAC sockperf: Latency Ping Pong redis: LPOP fftw: Stock - 2D FFT Size 1024 encode-mp3: WAV To MP3 bullet: 3000 Fall tjbench: Decompression Throughput polybench-c: Covariance Computation polybench-c: Correlation Computation fftw: Stock - 1D FFT Size 1024 polybench-c: 3 Matrix Multiplications tscp: AI Chess Performance sqlite: Timed SQLite Insertions bullet: Raytests bullet: Convex Trimesh bullet: Prim Trimesh bullet: 1000 Convex bullet: 136 Ragdolls bullet: 1000 Stack scimark2: Dense LU Matrix Factorization scimark2: Sparse Matrix Multiply scimark2: Monte Carlo scimark2: Jacobi Successive Over-Relaxation scimark2: Fast Fourier Transform GCC 8.2 RC1: -O3 -march=native AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native - PGO GCC 8.2 RC1: -O3 GCC 9.0 SVN: -O3 -march=native GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core 12.38 32.02 59730290 439.54 155.64 2071.78 408873.25 14374.13 36226.69 4 2.52 12497.20 100684 982.67 16205.32 188 110 185 154 209 50.80 7.65 1147009 6014866 1943.30 1691908.00 1375161.96 19.75 4610.63 1108304.54 1430638.45 12.63 6.42 1108268.08 6557.83 10.51 5.02 142.18 5.34 5.34 8518.67 3.58 851019 2.89 3.11 1.33 1.13 5.32 3.09 5.79 4831.75 2421.38 556.37 1687.64 219.38 2694.47 1372510 3.04 260.02 1427.38 242221.35 9213.45 28411.49 533 566.22 12463.50 98726 972.07 15210.47 6 23 29 18 46 57.85 192.26 1131696 5998759 1766.38 31.29 4656.40 12.87 6.88 6351.50 15.40 106.03 5.37 5.39 8289.10 3.66 729507 3.13 4301.99 2442.87 195.98 1676.49 214.57 12.29 32.44 61080869 3 2.51 12453.20 99762 852.63 49.05 1112448 6026655 1877.54 4617.23 5.42 5.33 3.35 919193 4865.24 2413.24 213.98 1676.15 219.09 12.36 32.09 59837512 441.58 166.70 2044.30 408599.13 13985.70 36248.65 4 2.84 12555.70 97917 998.89 15891.72 177 108 175 147 199 52.60 7.71 1113356 5989207 1874.09 1813896.31 1332541.96 19.67 4609.73 1187656.29 1517826.77 12.27 6.76 1143259.50 5835.90 11.13 5.06 141.74 5.39 5.40 7176.77 3.59 851283 2.88 3.18 1.36 1.12 5.52 3.14 6.02 4683.55 2473.98 561.85 1429.81 221.30 12.37 31.56 59201545 156.58 2015.18 401369.61 14307.54 36115.94 4 2.52 12377.60 97385 1007.74 15856.03 182 104 180 149 201 50.72 7.63 1111662 5876297 1940.65 1586367.58 1284845.13 19.88 4616.47 1133637.46 1423169.79 12.44 1152018.46 6531.37 10.50 5.04 145.66 5.41 5.38 8506.83 3.66 856576 2.89 3.12 1.33 1.13 5.29 3.08 5.78 4781.45 2458.23 556.37 1688.09 219.09 3263.94 261.52 1348.77 238819.60 8870.94 28285.32 560 561.50 12474.20 99362 1006.19 15244.02 6 23 30 18 46 57.00 192.14 1142302 5878800 1869.33 18.06 4647.80 12.87 6.79 6448.40 15.49 106.90 5.45 5.38 8345.20 3.61 756563 3.07 4869.99 2391.26 190.06 1678.45 216.86 OpenBenchmarking.org
VP9 libvpx Encoding vpxenc GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.7.0 vpxenc -O3 -march=native -O3 PGO -O3 -march=native 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 12.37 12.36 12.29 12.38 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
m-queens Time To Solve OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.1 Time To Solve GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 700 1400 2100 2800 3500 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 3263.94 31.56 32.09 32.44 2694.47 32.02 1. (CXX) g++ options: -fopenmp -O3 -march=native -O2
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 13M 26M 39M 52M 65M SE +/- 756265.41, N = 3 SE +/- 551889.61, N = 3 SE +/- 512120.63, N = 3 SE +/- 576746.80, N = 3 59201545 59837512 61080869 1372510 59730290 -march=native -march=native -fprofile-correction -march=native 1. (CXX) g++ options: -m64 -lpthread -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3a Phong Rendering With Soft-Shadow Mapping GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 100 200 300 400 500 SE +/- 1.61, N = 3 SE +/- 1.94, N = 3 441.58 3.04 439.54 -march=native 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++
Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 60 120 180 240 300 SE +/- 1.43, N = 3 SE +/- 0.68, N = 3 SE +/- 0.54, N = 3 261.52 156.58 166.70 260.02 155.64 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -O3 -pthread -lz -llzma
PostgreSQL pgbench Scaling: Buffer Test - Test: Single Thread - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 400 800 1200 1600 2000 SE +/- 37.36, N = 3 SE +/- 30.23, N = 5 SE +/- 12.52, N = 3 1348.77 2015.18 2044.30 1427.38 2071.78 -march=native -march=native -lpq -lpq -march=native -march=native -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 90K 180K 270K 360K 450K SE +/- 3319.69, N = 3 SE +/- 1716.32, N = 3 SE +/- 1289.79, N = 3 238819.60 401369.61 408599.13 242221.35 408873.25 -march=native -march=native -lpq -lpq -march=native -march=native -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Single Thread - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Only GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 3K 6K 9K 12K 15K SE +/- 152.49, N = 3 SE +/- 266.50, N = 3 SE +/- 113.33, N = 3 8870.94 14307.54 13985.70 9213.45 14374.13 -march=native -march=native -lpq -lpq -march=native -march=native -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 8K 16K 24K 32K 40K SE +/- 66.03, N = 3 SE +/- 81.46, N = 3 SE +/- 133.09, N = 3 28285.32 36115.94 36248.65 28411.49 36226.69 -march=native -march=native -lpq -lpq -march=native -march=native -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
Smallpt Global Illumination Renderer; 100 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 120 240 360 480 600 SE +/- 0.21, N = 6 560 4 4 3 533 4 -march=native -march=native 1. (CXX) g++ options: -fopenmp -O3
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 120 240 360 480 600 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 561.50 2.52 2.84 2.51 566.22 2.52 -march=native -march=native 1. (CC) gcc options: -lm -lpthread -O3
Tinymembench Standard Memset OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 2K 4K 6K 8K 10K 10034.00 9982.80 10037.70 9975.20 10015.50 10063.10 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O2 -O3 -lm
Tinymembench Standard Memcpy OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 3K 6K 9K 12K 15K 12474.20 12377.60 12555.70 12453.20 12463.50 12497.20 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O2 -O3 -lm
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 20K 40K 60K 80K 100K SE +/- 465.88, N = 3 SE +/- 586.49, N = 3 SE +/- 555.15, N = 3 SE +/- 715.28, N = 3 99362 97385 97917 99762 98726 100684 1. (CXX) g++ options: -pipe -lpthread
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 200 400 600 800 1000 SE +/- 4.57, N = 3 SE +/- 2.17, N = 3 SE +/- 0.73, N = 3 SE +/- 0.78, N = 3 1006.19 1007.74 998.89 852.63 972.07 982.67 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -mavx2
Apache Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 3K 6K 9K 12K 15K SE +/- 89.48, N = 3 SE +/- 188.72, N = 3 SE +/- 193.23, N = 3 15244.02 15856.03 15891.72 15210.47 16205.32 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -shared -fPIC -pthread -O3
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Sharpen GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 40 80 120 160 200 6 182 177 6 188 -march=native -march=native 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
GraphicsMagick Operation: Local Adaptive Thresholding OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Local Adaptive Thresholding GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 20 40 60 80 100 SE +/- 0.33, N = 3 23 104 108 23 110 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Resizing GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 40 80 120 160 200 SE +/- 1.00, N = 3 SE +/- 1.00, N = 3 SE +/- 1.20, N = 3 30 180 175 29 185 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
GraphicsMagick Operation: Blur OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Blur GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 30 60 90 120 150 18 149 147 18 154 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: HWB Color Space GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 50 100 150 200 250 SE +/- 0.58, N = 3 46 201 199 46 209 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
AOBench Size: 2048 x 2048 - Total Time OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 13 26 39 52 65 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.20, N = 3 SE +/- 0.01, N = 3 57.00 50.72 52.60 49.05 57.85 50.80 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -lm -O3
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 40 80 120 160 200 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 192.14 7.63 7.71 192.26 7.65 -march=native -march=native 1. (CC) gcc options: -O3 -pthread -lhmmer -lsquid -lm
ebizzy OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 200K 400K 600K 800K 1000K SE +/- 23788.74, N = 6 SE +/- 18183.13, N = 4 SE +/- 20234.15, N = 6 SE +/- 28582.82, N = 6 1142302 1111662 1113356 1112448 1131696 1147009 -fprofile-correction 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Crafty Elapsed Time OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 13470.58, N = 3 SE +/- 13110.39, N = 3 SE +/- 11021.16, N = 3 SE +/- 5146.22, N = 3 5878800 5876297 5989207 6026655 5998759 6014866 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 400 800 1200 1600 2000 SE +/- 4.00, N = 3 SE +/- 5.53, N = 3 SE +/- 3.46, N = 3 SE +/- 5.96, N = 3 1869.33 1940.65 1874.09 1877.54 1766.38 1943.30 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
Redis Test: GET GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET -O3 -march=native -O3 400K 800K 1200K 1600K 2000K SE +/- 25099.79, N = 6 SE +/- 34086.99, N = 6 SE +/- 37527.46, N = 6 1586367.58 1813896.31 1691908.00 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Redis Test: SET GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET -O3 -march=native -O3 300K 600K 900K 1200K 1500K SE +/- 31380.33, N = 6 SE +/- 23587.86, N = 6 SE +/- 13759.71, N = 3 1284845.13 1332541.96 1375161.96 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Sockperf Test: Latency Under Load OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 7 14 21 28 35 SE +/- 0.24, N = 8 SE +/- 0.31, N = 10 SE +/- 0.30, N = 10 18.06 19.88 19.67 31.29 19.75 -march=native -march=native -march=native -march=native 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.0f RSA 4096-bit Performance GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1000 2000 3000 4000 5000 SE +/- 23.13, N = 3 SE +/- 22.71, N = 3 SE +/- 23.93, N = 3 SE +/- 21.56, N = 3 4647.80 4616.47 4609.73 4617.23 4656.40 4610.63 -lssl -lssl -lssl 1. (CC) gcc options: -O3 -pthread -m64 -lcrypto -ldl
Redis Test: LPUSH GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH -O3 -march=native -O3 300K 600K 900K 1200K 1500K SE +/- 12639.17, N = 3 SE +/- 2157.90, N = 3 SE +/- 18347.41, N = 6 1133637.46 1187656.29 1108304.54 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Redis Test: SADD GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD -O3 -march=native -O3 300K 600K 900K 1200K 1500K SE +/- 3767.82, N = 3 SE +/- 35661.81, N = 6 SE +/- 4081.67, N = 3 1423169.79 1517826.77 1430638.45 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 3 6 9 12 15 SE +/- 0.01, N = 5 SE +/- 0.12, N = 5 SE +/- 0.19, N = 5 12.87 12.44 12.27 12.87 12.63 -march=native -march=native -march=native -march=native 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
Sockperf Test: Latency Ping Pong OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 2 4 6 8 10 SE +/- 0.07, N = 5 SE +/- 0.11, N = 10 6.79 6.76 6.88 6.42 -march=native -march=native -march=native 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Redis Test: LPOP GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP -O3 -march=native -O3 200K 400K 600K 800K 1000K SE +/- 21693.88, N = 3 SE +/- 15264.52, N = 3 SE +/- 4079.04, N = 3 1152018.46 1143259.50 1108268.08 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
FFTW Build: Stock - Size: 2D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1400 2800 4200 5600 7000 SE +/- 16.29, N = 3 SE +/- 3.18, N = 3 SE +/- 73.40, N = 3 6448.40 6531.37 5835.90 6351.50 6557.83 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -pthread -O3 -lm
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 15.49 10.50 11.13 15.40 10.51 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -O3 -lm
Bullet Physics Engine Test: 3000 Fall GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall -O3 -march=native -O3 1.1385 2.277 3.4155 4.554 5.6925 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 5.04 5.06 5.02 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 106.90 145.66 141.74 106.03 142.18 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -O3 -lm
PolyBench-C Test: Covariance Computation OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Covariance Computation GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1.2263 2.4526 3.6789 4.9052 6.1315 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 5.45 5.41 5.39 5.42 5.37 5.34 -fprofile-correction 1. (CC) gcc options: -O3 -march=native
PolyBench-C Test: Correlation Computation OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Correlation Computation GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1.215 2.43 3.645 4.86 6.075 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 5.38 5.38 5.40 5.33 5.39 5.34 -fprofile-correction 1. (CC) gcc options: -O3 -march=native
FFTW Build: Stock - Size: 1D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 2K 4K 6K 8K 10K SE +/- 1.83, N = 3 SE +/- 8.90, N = 3 SE +/- 18.71, N = 3 8345.20 8506.83 7176.77 8289.10 8518.67 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -pthread -O3 -lm
PolyBench-C Test: 3 Matrix Multiplications OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: 3 Matrix Multiplications GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 0.8235 1.647 2.4705 3.294 4.1175 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 3.61 3.66 3.59 3.35 3.66 3.58 -fprofile-correction 1. (CC) gcc options: -O3 -march=native
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 200K 400K 600K 800K 1000K SE +/- 421.22, N = 5 SE +/- 833.02, N = 5 SE +/- 1018.49, N = 5 SE +/- 766.17, N = 5 756563 856576 851283 919193 729507 851019 -fprofile-correction 1. (CC) gcc options: -O3 -march=native
SQLite Timed SQLite Insertions OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.22 Timed SQLite Insertions GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 0.7043 1.4086 2.1129 2.8172 3.5215 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 3.07 2.89 2.88 3.13 2.89 -march=native -march=native -march=native -march=native 1. (CC) gcc options: -O3 -lz -ldl -lpthread
Bullet Physics Engine Test: Raytests GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Raytests -O3 -march=native -O3 0.7155 1.431 2.1465 2.862 3.5775 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.12 3.18 3.11 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Bullet Physics Engine Test: Convex Trimesh GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh -O3 -march=native -O3 0.306 0.612 0.918 1.224 1.53 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.33 1.36 1.33 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Bullet Physics Engine Test: Prim Trimesh GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh -O3 -march=native -O3 0.2543 0.5086 0.7629 1.0172 1.2715 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.13 1.12 1.13 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Bullet Physics Engine Test: 1000 Convex GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex -O3 -march=native -O3 1.242 2.484 3.726 4.968 6.21 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.29 5.52 5.32 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Bullet Physics Engine Test: 136 Ragdolls GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls -O3 -march=native -O3 0.7065 1.413 2.1195 2.826 3.5325 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 3.08 3.14 3.09 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Bullet Physics Engine Test: 1000 Stack GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack -O3 -march=native -O3 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.78 6.02 5.79 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 1000 2000 3000 4000 5000 SE +/- 25.52, N = 3 SE +/- 27.07, N = 3 SE +/- 17.02, N = 3 SE +/- 20.82, N = 3 4869.99 4781.45 4683.55 4865.24 4301.99 4831.75 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 500 1000 1500 2000 2500 SE +/- 13.99, N = 3 SE +/- 1.01, N = 3 SE +/- 10.35, N = 3 SE +/- 11.03, N = 3 2391.26 2458.23 2473.98 2413.24 2442.87 2421.38 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 120 240 360 480 600 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.25, N = 3 SE +/- 0.10, N = 3 190.06 556.37 561.85 213.98 195.98 556.37 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 400 800 1200 1600 2000 SE +/- 0.72, N = 3 SE +/- 0.16, N = 3 SE +/- 0.26, N = 3 SE +/- 0.06, N = 3 1678.45 1688.09 1429.81 1676.15 1676.49 1687.64 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform GCC 9.0.0 20180720 - AMD EPYC 7601 32-Core GCC 9.0 SVN: -O3 -march=native GCC 8.2 RC1: -O3 GCC 8.2 RC1: -O3 -march=native - PGO AMD EPYC 7601 32-Core GCC 8.2 RC1: -O3 -march=native 50 100 150 200 250 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.10, N = 3 216.86 219.09 221.30 219.09 214.57 219.38 -march=native -march=native -march=native -fprofile-correction -march=native -march=native 1. (CC) gcc options: -O3 -lm
Phoronix Test Suite v10.8.4