AMD Zen 2 GCC compiler benchmarks on Ubuntu Linux. Tests by Michael Larabel for a future article.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1907117-HV-ZNVER2GCC44 Znver2 GCC9 Compiler Tests - Phoronix Test Suite Znver2 GCC9 Compiler Tests AMD Zen 2 GCC compiler benchmarks on Ubuntu Linux. Tests by Michael Larabel for a future article.
HTML result view exported from: https://openbenchmarking.org/result/1907117-HV-ZNVER2GCC44&grs&rdt .
Znver2 GCC9 Compiler Tests Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads) ASUS ROG CROSSHAIR VIII HERO (WI-FI) (0066 BIOS) AMD Device 1480 16384MB 2000GB Force MP600 Sapphire AMD Baffin [Polaris11] 4GB (1300/1750MHz) AMD Device aae0 ASUS VP28U Realtek Device 8125 + Intel I211 + Intel Device 2723 Ubuntu 18.04 5.2.0-999-generic (x86_64) 20190703 GNOME Shell 3.28.3 X Server 1.20.1 modesetting 1.20.1 4.5 Mesa 18.2.2 (LLVM 7.0.0) GCC 9.1.0 ext4 3840x2160 OpenBenchmarking.org Environment Details - -O3 -march=znver2: CXXFLAGS=-O3-march=znver2 CFLAGS=-O3-march=znver2 - -O3 -march=znver1: CXXFLAGS=-O3-march=znver1 CFLAGS=-O3-march=znver1 - -O3 -march=x86-64: CXXFLAGS=-O3-march=x86-64 CFLAGS=-O3-march=x86-64 Compiler Details - --disable-multilib --enable-checking=release Processor Details - Scaling Governor: acpi-cpufreq ondemand Python Details - Python 2.7.15+ + Python 3.6.8 Security Details - l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled
Znver2 GCC9 Compiler Tests scimark2: Dense LU Matrix Factorization john-the-ripper: Blowfish scimark2: Composite fftw: Stock - 1D FFT Size 4096 scimark2: Fast Fourier Transform fftw: Stock - 2D FFT Size 4096 scimark2: Jacobi Successive Over-Relaxation c-ray: Total Time - 4K, 16 Rays Per Pixel fftw: Float + SSE - 1D FFT Size 4096 graphics-magick: Sharpen bullet: 1000 Stack graphics-magick: Rotate bullet: 136 Ragdolls graphics-magick: Resizing scimark2: Monte Carlo aobench: 2048 x 2048 - Total Time scimark2: Sparse Matrix Multiply encode-flac: WAV To FLAC bullet: 3000 Fall mcperf: Get bullet: 1000 Convex tscp: AI Chess Performance bullet: Raytests cpp-perf-bench: Function Objects fftw: Float + SSE - 2D FFT Size 4096 pgbench: Buffer Test - Normal Load - Read Write encode-mp3: WAV To MP3 mkl-dnn: Convolution Batch conv_alexnet - f32 mkl-dnn: Deconvolution Batch deconv_1d - f32 build-llvm: Time To Compile x264: H.264 Video Encoding stockfish: Total Time build-php: Time To Compile cpp-perf-bench: Math Library vpxenc: vpxenc VP9 1080p Video Encode himeno: Poisson Pressure Solver x265: H.265 1080p Video Encoding pgbench: Buffer Test - Normal Load - Read Only redis: SET compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 compress-7zip: Compress Speed Test mcperf: Set redis: GET mkl-dnn: IP Batch 1D - f32 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 11431.93 20232 3700.64 11231 274.11 8001.67 2422.10 39.46 56652 194 3.79 276 2.05 285 799.07 34.64 3575.96 8.12 3.23 110755.89 3.59 1321681 2.06 14.56 19960 30044.32 7.04 2524.93 216.70 286.56 140.96 39471726 53.41 309.02 175.34 1347.96 52.91 382751.07 2089609.47 25.21 78562 69121.64 3090850.69 158.46 8631.93 28221 3128.65 11448 260.12 7660.90 2291.62 39.42 51757 193 3.98 260 2.13 279 757.56 35.14 3702.03 8.15 3.36 112447.77 3.73 1372148 2.13 14.99 19405 29380.50 6.98 2562.89 219.29 284.24 141.82 39908751 53.44 311.77 174.40 1345.95 53.15 383329.30 2080013.12 25.09 78412 60810.38 3126726.23 159.95 6959.69 28401 2786.33 9534.70 297.13 7039.43 2145.57 43.20 179 4.08 261 2.17 270 766.81 36.49 3762.43 7.75 3.37 107814.85 3.70 1333926 2.09 14.90 30143.80 7.16 2512.04 221.00 281.90 143.27 39537930 52.89 312.01 175.99 1336.77 53.33 385510.95 2074827.24 25.22 78655 60770.80 3026826.32 152.36 OpenBenchmarking.org
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 2K 4K 6K 8K 10K SE +/- 10.78, N = 3 SE +/- 25.99, N = 3 SE +/- 73.09, N = 3 11431.93 8631.93 6959.69 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 6K 12K 18K 24K 30K SE +/- 50.35, N = 3 SE +/- 51.72, N = 3 SE +/- 61.75, N = 3 20232 28221 28401 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 800 1600 2400 3200 4000 SE +/- 7.82, N = 3 SE +/- 7.70, N = 3 SE +/- 27.49, N = 3 3700.64 3128.65 2786.33 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
FFTW Build: Stock - Size: 1D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 4096 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 2K 4K 6K 8K 10K SE +/- 123.02, N = 3 SE +/- 53.59, N = 3 SE +/- 11.52, N = 3 11231.00 11448.00 9534.70 -march=znver2 -march=znver1 1. (CC) gcc options: -pthread -O3 -lm
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 60 120 180 240 300 SE +/- 0.15, N = 3 SE +/- 0.04, N = 3 SE +/- 3.16, N = 3 274.11 260.12 297.13 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
FFTW Build: Stock - Size: 2D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 2K 4K 6K 8K 10K SE +/- 28.05, N = 3 SE +/- 34.96, N = 3 SE +/- 7.75, N = 3 8001.67 7660.90 7039.43 -march=znver2 -march=znver1 1. (CC) gcc options: -pthread -O3 -lm
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 500 1000 1500 2000 2500 SE +/- 0.28, N = 3 SE +/- 0.13, N = 3 SE +/- 20.09, N = 3 2422.10 2291.62 2145.57 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 10 20 30 40 50 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 39.46 39.42 43.20 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -lm -lpthread -O3
FFTW Build: Float + SSE - Size: 1D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 -O3 -march=znver2 -O3 -march=znver1 12K 24K 36K 48K 60K SE +/- 647.09, N = 3 SE +/- 455.36, N = 3 56652 51757 -march=znver2 -march=znver1 1. (CC) gcc options: -pthread -O3 -lm
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Sharpen -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 40 80 120 160 200 SE +/- 0.58, N = 3 194 193 179 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
Bullet Physics Engine Test: 1000 Stack OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 0.918 1.836 2.754 3.672 4.59 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 3.79 3.98 4.08 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Rotate -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 60 120 180 240 300 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 276 260 261 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
Bullet Physics Engine Test: 136 Ragdolls OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 0.4883 0.9766 1.4649 1.9532 2.4415 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 2.05 2.13 2.17 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Resizing -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 60 120 180 240 300 SE +/- 1.53, N = 3 285 279 270 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 200 400 600 800 1000 SE +/- 1.53, N = 3 SE +/- 0.16, N = 3 SE +/- 7.83, N = 3 799.07 757.56 766.81 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
AOBench Size: 2048 x 2048 - Total Time OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 8 16 24 32 40 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 SE +/- 0.32, N = 3 34.64 35.14 36.49 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -lm -O3
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 800 1600 2400 3200 4000 SE +/- 48.59, N = 3 SE +/- 48.59, N = 3 SE +/- 48.28, N = 3 3575.96 3702.03 3762.43 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 2 4 6 8 10 SE +/- 0.06, N = 5 SE +/- 0.04, N = 5 SE +/- 0.04, N = 5 8.12 8.15 7.75 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
Bullet Physics Engine Test: 3000 Fall OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 0.7583 1.5166 2.2749 3.0332 3.7915 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 3.23 3.36 3.37 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Memcached mcperf Method: Get OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Get -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 20K 40K 60K 80K 100K SE +/- 1084.23, N = 15 SE +/- 1283.75, N = 15 SE +/- 879.11, N = 15 110755.89 112447.77 107814.85 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm -rdynamic
Bullet Physics Engine Test: 1000 Convex OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 0.8393 1.6786 2.5179 3.3572 4.1965 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 3.59 3.73 3.70 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 300K 600K 900K 1200K 1500K SE +/- 8868.66, N = 5 SE +/- 1369.78, N = 5 SE +/- 10274.76, N = 5 1321681 1372148 1333926 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -march=native
Bullet Physics Engine Test: Raytests OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Raytests -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 0.4793 0.9586 1.4379 1.9172 2.3965 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 2.06 2.13 2.09 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
CppPerformanceBenchmarks Test: Function Objects OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 4 8 12 16 20 SE +/- 0.16, N = 3 SE +/- 0.01, N = 3 SE +/- 0.16, N = 3 14.56 14.99 14.90 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -std=c++11
FFTW Build: Float + SSE - Size: 2D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 -O3 -march=znver2 -O3 -march=znver1 4K 8K 12K 16K 20K SE +/- 71.31, N = 3 SE +/- 83.01, N = 3 19960 19405 -march=znver2 -march=znver1 1. (CC) gcc options: -pthread -O3 -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 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 6K 12K 18K 24K 30K SE +/- 197.60, N = 3 SE +/- 361.55, N = 15 SE +/- 169.86, N = 3 30044.32 29380.50 30143.80 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 2 4 6 8 10 SE +/- 0.12, N = 3 SE +/- 0.09, N = 4 SE +/- 0.07, N = 3 7.04 6.98 7.16 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lncurses -lm
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: f32 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 500 1000 1500 2000 2500 SE +/- 12.02, N = 3 SE +/- 19.59, N = 3 SE +/- 11.23, N = 3 2524.93 2562.89 2512.04 -march=znver2 - MIN: 2478.96 -march=znver1 - MIN: 2476.4 -march=x86-64 - MIN: 2466.12 1. (CXX) g++ options: -O3 -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: f32 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 50 100 150 200 250 SE +/- 2.63, N = 6 SE +/- 1.89, N = 11 SE +/- 2.51, N = 3 216.70 219.29 221.00 -march=znver2 - MIN: 202.77 -march=znver1 - MIN: 203.58 -march=x86-64 - MIN: 202.64 1. (CXX) g++ options: -O3 -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 6.0.1 Time To Compile -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 60 120 180 240 300 286.56 284.24 281.90
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 30 60 90 120 150 SE +/- 0.93, N = 3 SE +/- 0.82, N = 3 SE +/- 0.64, N = 3 140.96 141.82 143.27 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 9M 18M 27M 36M 45M SE +/- 265193.75, N = 3 SE +/- 208989.61, N = 3 SE +/- 75524.60, N = 3 39471726 39908751 39537930 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -m64 -lpthread -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.1.9 Time To Compile -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 12 24 36 48 60 SE +/- 0.20, N = 3 SE +/- 0.10, N = 3 SE +/- 0.34, N = 3 53.41 53.44 52.89 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -pedantic -ldl -lz -lm
CppPerformanceBenchmarks Test: Math Library OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Math Library -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 70 140 210 280 350 SE +/- 1.69, N = 3 SE +/- 5.00, N = 3 SE +/- 4.58, N = 4 309.02 311.77 312.01 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -std=c++11
VP9 libvpx Encoding vpxenc VP9 1080p Video Encode OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 40 80 120 160 200 SE +/- 0.55, N = 3 SE +/- 0.98, N = 3 SE +/- 0.86, N = 3 175.34 174.40 175.99 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 300 600 900 1200 1500 SE +/- 12.91, N = 3 SE +/- 22.06, N = 3 SE +/- 13.46, N = 3 1347.96 1345.95 1336.77 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -mavx2
x265 H.265 1080p Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 12 24 36 48 60 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 52.91 53.15 53.33 -march=znver2 -march=znver1 -march=x86-64 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
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 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 80K 160K 240K 320K 400K SE +/- 738.26, N = 3 SE +/- 539.80, N = 3 SE +/- 837.27, N = 3 382751.07 383329.30 385510.95 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 400K 800K 1200K 1600K 2000K SE +/- 26820.33, N = 15 SE +/- 26581.97, N = 15 SE +/- 28039.49, N = 15 2089609.47 2080013.12 2074827.24 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 6 12 18 24 30 SE +/- 0.10, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 25.21 25.09 25.22 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -pthread -fvisibility=hidden -O3
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 20K 40K 60K 80K 100K SE +/- 205.86, N = 3 SE +/- 193.52, N = 3 SE +/- 419.64, N = 3 78562 78412 78655 1. (CXX) g++ options: -pipe -lpthread
Memcached mcperf Method: Set OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Set -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 15K 30K 45K 60K 75K SE +/- 4868.10, N = 15 SE +/- 1488.06, N = 15 SE +/- 807.63, N = 4 69121.64 60810.38 60770.80 -march=znver2 -march=znver1 -march=x86-64 1. (CC) gcc options: -O3 -lm -rdynamic
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 700K 1400K 2100K 2800K 3500K SE +/- 61045.92, N = 12 SE +/- 62689.04, N = 13 SE +/- 58033.83, N = 15 3090850.69 3126726.23 3026826.32 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
MKL-DNN Harness: IP Batch 1D - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 -O3 -march=znver2 -O3 -march=znver1 -O3 -march=x86-64 40 80 120 160 200 SE +/- 2.44, N = 15 SE +/- 2.27, N = 15 SE +/- 3.65, N = 15 158.46 159.95 152.36 -march=znver2 - MIN: 112.01 -march=znver1 - MIN: 108.37 -march=x86-64 - MIN: 109.43 1. (CXX) g++ options: -O3 -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Phoronix Test Suite v10.8.4