AMD TR 2920X Linux AMD Ryzen Threadripper 2920X 12-Core testing with a Gigabyte X399 AORUS Gaming 7 (F11e BIOS) and eVGA NVIDIA NV137 4GB on Ubuntu 18.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1906246-PTS-AMDTR29220&gru .
AMD TR 2920X Linux Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution Threadripper 2920X AMD Ryzen Threadripper 2920X 12-Core @ 3.50GHz (12 Cores / 24 Threads) Gigabyte X399 AORUS Gaming 7 (F11e BIOS) AMD 17h 16384MB 240GB Force MP510 + 120GB Force MP500 eVGA NVIDIA NV137 4GB Realtek ALC1220 ASUS VP28U Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 Ubuntu 18.04 4.19.0-041900-generic (x86_64) GNOME Shell 3.28.3 X Server 1.19.6 modesetting 1.19.6 4.3 Mesa 18.0.5 GCC 7.3.0 ext4 3840x2160 OpenBenchmarking.org - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq ondemand - OpenJDK Runtime Environment (build 10.0.2+13-Ubuntu-1ubuntu0.18.04.4) - Python 2.7.15rc1 + Python 3.6.7 - l1tf: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB
AMD TR 2920X Linux svt-av1: 1080p 8-bit YUV To AV1 Video Encode vpxenc: vpxenc VP9 1080p Video Encode indigobench: Bedroom indigobench: Supercar john-the-ripper: Blowfish john-the-ripper: MD5 nginx: Static Web Page Serving apache: Static Web Page Serving selenium: Jetstream - Firefox pgbench: Buffer Test - Normal Load - Read Only pgbench: Buffer Test - Normal Load - Read Write pgbench: Buffer Test - Single Thread - Read Only pgbench: Buffer Test - Single Thread - Read Write pgbench: Buffer Test - Heavy Contention - Read Only pgbench: Buffer Test - Heavy Contention - Read Write apache-siege: 10 apache-siege: 50 apache-siege: 100 apache-siege: 200 apache-siege: 250 ctx-clock: Context Switch Time renaissance: Scala Dotty renaissance: Apache Spark ALS renaissance: Apache Spark Bayes renaissance: Savina Reactors.IO renaissance: Apache Spark PageRank renaissance: In-Memory Database Shootout renaissance: Akka Unbalanced Cobwebbed Tree mkl-dnn: IP Batch 1D - f32 mkl-dnn: IP Batch All - f32 mkl-dnn: IP Batch 1D - u8s8u8s32 mkl-dnn: IP Batch 1D - u8s8f32s32 mkl-dnn: IP Batch All - u8s8u8s32 mkl-dnn: IP Batch All - u8s8f32s32 mkl-dnn: Convolution Batch conv_3d - f32 mkl-dnn: Convolution Batch conv_all - f32 mkl-dnn: Deconvolution Batch deconv_1d - f32 mkl-dnn: Deconvolution Batch deconv_3d - f32 mkl-dnn: Convolution Batch conv_alexnet - f32 mkl-dnn: Deconvolution Batch deconv_all - f32 mkl-dnn: Convolution Batch conv_3d - u8s8u8s32 mkl-dnn: Convolution Batch conv_3d - u8s8f32s32 mkl-dnn: Convolution Batch conv_all - u8s8u8s32 mkl-dnn: Convolution Batch conv_all - u8s8f32s32 mkl-dnn: Convolution Batch conv_googlenet_v3 - f32 mkl-dnn: Deconvolution Batch deconv_1d - u8s8u8s32 mkl-dnn: Deconvolution Batch deconv_3d - u8s8u8s32 mkl-dnn: Convolution Batch conv_alexnet - u8s8u8s32 mkl-dnn: Deconvolution Batch deconv_1d - u8s8f32s32 mkl-dnn: Deconvolution Batch deconv_3d - u8s8f32s32 mkl-dnn: Deconvolution Batch deconv_all - u8s8u8s32 mkl-dnn: Convolution Batch conv_alexnet - u8s8f32s32 mkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8u8s32 mkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32s32 core-latency: Average Latency Between CPU Cores t-test1: 1 t-test1: 2 cp2k: Fayalite-FIST Data primesieve: 1e12 Prime Number Generation compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 tachyon: Total Time gimp: resize gimp: rotate gimp: auto-levels gimp: unsharp-mask Threadripper 2920X 33.80 138.59 1.70 3.63 27037 855833 30180.32 25252.94 185.38 432356.73 21146.40 29528.05 1136.09 436923.83 20648.68 25041.98 24888.75 22670.83 22611.65 35673.52 175 6522.49 6454.55 4308.11 22180.48 20976.97 7366.48 14843.76 14.36 174.55 54.74 54.74 608.89 614.60 23.33 3917.84 22.95 9.50 484.67 5754.33 7644.85 7638.65 39375.00 39053.70 215.90 3436.87 6117.86 4225.45 3423.59 6095.41 29785.53 4217.91 1847.85 1814.92 357.81 28.17 9.54 712.14 17.17 26.88 3.78 7.34 15.89 15.42 34.75 OpenBenchmarking.org
SVT-AV1 1080p 8-bit YUV To AV1 Video Encode OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.5 1080p 8-bit YUV To AV1 Video Encode Threadripper 2920X 8 16 24 32 40 SE +/- 0.14, N = 3 33.80 1. (CXX) g++ options: -O3 -pie -lpthread -lm
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 Threadripper 2920X 30 60 90 120 150 SE +/- 0.28, N = 3 138.59 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
IndigoBench Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Bedroom Threadripper 2920X 0.3825 0.765 1.1475 1.53 1.9125 SE +/- 0.00, N = 3 1.70
IndigoBench Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Supercar Threadripper 2920X 0.8168 1.6336 2.4504 3.2672 4.084 SE +/- 0.01, N = 3 3.63
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish Threadripper 2920X 6K 12K 18K 24K 30K SE +/- 24.69, N = 3 27037 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
John The Ripper Test: MD5 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 Threadripper 2920X 200K 400K 600K 800K 1000K SE +/- 3242.06, N = 3 855833 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
NGINX Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving Threadripper 2920X 6K 12K 18K 24K 30K SE +/- 333.41, N = 10 30180.32 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
Apache Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 14.99, N = 3 25252.94 1. (CC) gcc options: -shared -fPIC -O2 -pthread
Selenium Benchmark: Jetstream - Browser: Firefox OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream - Browser: Firefox Threadripper 2920X 40 80 120 160 200 SE +/- 0.67, N = 3 185.38 1. firefox 67.0.3
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 Threadripper 2920X 90K 180K 270K 360K 450K SE +/- 2476.73, N = 3 432356.73 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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 Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 532.34, N = 15 21146.40 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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 Threadripper 2920X 6K 12K 18K 24K 30K SE +/- 337.86, N = 9 29528.05 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
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 Threadripper 2920X 200 400 600 800 1000 SE +/- 51.66, N = 13 1136.09 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only Threadripper 2920X 90K 180K 270K 360K 450K SE +/- 3007.54, N = 3 436923.83 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write Threadripper 2920X 4K 8K 12K 16K 20K SE +/- 515.85, N = 15 20648.68 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Apache Siege Concurrent Users: 10 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 10 Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 55.24, N = 3 25041.98 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 50 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 95.97, N = 3 24888.75 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 100 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 71.95, N = 2 22670.83 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 200 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 Threadripper 2920X 5K 10K 15K 20K 25K 22611.65 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 250 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 Threadripper 2920X 8K 16K 24K 32K 40K 35673.52 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
ctx_clock Context Switch Time OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time Threadripper 2920X 40 80 120 160 200 175
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Scala Dotty Threadripper 2920X 1400 2800 4200 5600 7000 SE +/- 49.97, N = 8 6522.49
Renaissance Test: Apache Spark ALS OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark ALS Threadripper 2920X 1400 2800 4200 5600 7000 SE +/- 170.82, N = 40 6454.55
Renaissance Test: Apache Spark Bayes OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark Bayes Threadripper 2920X 900 1800 2700 3600 4500 SE +/- 27.05, N = 40 4308.11
Renaissance Test: Savina Reactors.IO OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Savina Reactors.IO Threadripper 2920X 5K 10K 15K 20K 25K SE +/- 313.15, N = 40 22180.48
Renaissance Test: Apache Spark PageRank OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark PageRank Threadripper 2920X 4K 8K 12K 16K 20K SE +/- 134.54, N = 8 20976.97
Renaissance Test: In-Memory Database Shootout OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: In-Memory Database Shootout Threadripper 2920X 1600 3200 4800 6400 8000 SE +/- 59.43, N = 8 7366.48
Renaissance Test: Akka Unbalanced Cobwebbed Tree OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Akka Unbalanced Cobwebbed Tree Threadripper 2920X 3K 6K 9K 12K 15K SE +/- 166.13, N = 40 14843.76
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 Threadripper 2920X 4 8 12 16 20 SE +/- 0.02, N = 3 14.36 MIN: 14.1 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch All - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: f32 Threadripper 2920X 40 80 120 160 200 SE +/- 0.72, N = 3 174.55 MIN: 171.36 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch 1D - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8u8s32 Threadripper 2920X 12 24 36 48 60 SE +/- 0.76, N = 3 54.74 MIN: 52.82 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch 1D - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8f32s32 Threadripper 2920X 12 24 36 48 60 SE +/- 0.60, N = 3 54.74 MIN: 52.78 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch All - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8u8s32 Threadripper 2920X 130 260 390 520 650 SE +/- 1.25, N = 3 608.89 MIN: 594.16 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch All - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8f32s32 Threadripper 2920X 130 260 390 520 650 SE +/- 6.56, N = 3 614.60 MIN: 598.1 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: f32 Threadripper 2920X 6 12 18 24 30 SE +/- 0.06, N = 3 23.33 MIN: 22.6 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: f32 Threadripper 2920X 800 1600 2400 3200 4000 SE +/- 10.64, N = 3 3917.84 MIN: 3840.16 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -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 Threadripper 2920X 5 10 15 20 25 SE +/- 0.03, N = 3 22.95 MIN: 22.56 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: f32 Threadripper 2920X 3 6 9 12 15 SE +/- 0.16, N = 3 9.50 MIN: 9.05 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
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 Threadripper 2920X 100 200 300 400 500 SE +/- 1.21, N = 3 484.67 MIN: 480.61 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_all - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_all - Data Type: f32 Threadripper 2920X 1200 2400 3600 4800 6000 SE +/- 17.67, N = 3 5754.33 MIN: 5617.69 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32 Threadripper 2920X 1600 3200 4800 6400 8000 SE +/- 16.61, N = 3 7644.85 MIN: 7602.18 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32 Threadripper 2920X 1600 3200 4800 6400 8000 SE +/- 4.05, N = 3 7638.65 MIN: 7622.75 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: u8s8u8s32 Threadripper 2920X 8K 16K 24K 32K 40K SE +/- 171.16, N = 3 39375.00 MIN: 38813.3 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: u8s8f32s32 Threadripper 2920X 8K 16K 24K 32K 40K SE +/- 84.29, N = 3 39053.70 MIN: 38722.6 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 Threadripper 2920X 50 100 150 200 250 SE +/- 1.74, N = 3 215.90 MIN: 209.78 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8u8s32 Threadripper 2920X 700 1400 2100 2800 3500 SE +/- 6.94, N = 3 3436.87 MIN: 3419.21 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8u8s32 Threadripper 2920X 1300 2600 3900 5200 6500 SE +/- 9.73, N = 3 6117.86 MIN: 6083.12 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: u8s8u8s32 Threadripper 2920X 900 1800 2700 3600 4500 SE +/- 41.59, N = 3 4225.45 MIN: 4172.42 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32s32 Threadripper 2920X 700 1400 2100 2800 3500 SE +/- 3.55, N = 3 3423.59 MIN: 3411.48 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32s32 Threadripper 2920X 1300 2600 3900 5200 6500 SE +/- 8.40, N = 3 6095.41 MIN: 6069.02 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_all - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_all - Data Type: u8s8u8s32 Threadripper 2920X 6K 12K 18K 24K 30K SE +/- 19.95, N = 3 29785.53 MIN: 29521.7 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32s32 Threadripper 2920X 900 1800 2700 3600 4500 SE +/- 19.37, N = 3 4217.91 MIN: 4172.95 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8u8s32 Threadripper 2920X 400 800 1200 1600 2000 SE +/- 2.79, N = 3 1847.85 MIN: 1826.23 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32s32 Threadripper 2920X 400 800 1200 1600 2000 SE +/- 1.20, N = 3 1814.92 MIN: 1794.85 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
Core-Latency Average Latency Between CPU Cores OpenBenchmarking.org ns, Fewer Is Better Core-Latency Average Latency Between CPU Cores Threadripper 2920X 80 160 240 320 400 357.81 MIN: 35.88 / MAX: 555.49 1. (CXX) g++ options: -std=c++11 -pthread -O3
t-test1 Threads: 1 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 Threadripper 2920X 7 14 21 28 35 SE +/- 0.28, N = 3 28.17 1. (CC) gcc options: -pthread
t-test1 Threads: 2 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 Threadripper 2920X 3 6 9 12 15 SE +/- 0.05, N = 3 9.54 1. (CC) gcc options: -pthread
CP2K Molecular Dynamics Fayalite-FIST Data OpenBenchmarking.org Seconds, Fewer Is Better CP2K Molecular Dynamics 6.1 Fayalite-FIST Data Threadripper 2920X 150 300 450 600 750 712.14
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.4 1e12 Prime Number Generation Threadripper 2920X 4 8 12 16 20 SE +/- 0.03, N = 3 17.17 1. (CXX) g++ options: -O3 -lpthread
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 Threadripper 2920X 6 12 18 24 30 SE +/- 0.09, N = 3 26.88 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Tachyon Total Time OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.98.9 Total Time Threadripper 2920X 0.8505 1.701 2.5515 3.402 4.2525 SE +/- 0.01, N = 3 3.78 1. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
GIMP Test: resize OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.8.22 Test: resize Threadripper 2920X 2 4 6 8 10 SE +/- 0.07, N = 3 7.34
GIMP Test: rotate OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.8.22 Test: rotate Threadripper 2920X 4 8 12 16 20 SE +/- 0.03, N = 3 15.89
GIMP Test: auto-levels OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.8.22 Test: auto-levels Threadripper 2920X 4 8 12 16 20 SE +/- 0.09, N = 3 15.42
GIMP Test: unsharp-mask OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.8.22 Test: unsharp-mask Threadripper 2920X 8 16 24 32 40 SE +/- 0.10, N = 3 34.75
Phoronix Test Suite v10.8.4