Xeon Gold Benchmark Fun

2 x Intel Xeon Gold 6138 testing with a TYAN S7106 (V2.00.B20 BIOS) and llvmpipe 93GB on Ubuntu 18.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/1910060-AS-XEONGOLDB99.

Xeon Gold Benchmark FunProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution2 x Intel Xeon Gold 61382 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads)TYAN S7106 (V2.00.B20 BIOS)Intel Sky Lake-E DMI3 Registers96256MB500GB Samsung SSD 860llvmpipe 93GBVE2282 x Intel I210 + 2 x QLogic cLOM8214 1/10GbEUbuntu 18.045.0.0-29-generic (x86_64)GNOME Shell 3.28.4X Server 1.20.4modesetting 1.20.43.3 Mesa 19.0.8 (LLVM 8.0 256 bits)GCC 7.4.0ext41920x1080OpenBenchmarking.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: intel_pstate powersave- 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling

Xeon Gold Benchmark Funlczero: BLASlczero: Randlibgav1: Chimera 1080plibgav1: Summer Nature 4Klibgav1: Summer Nature 1080plibgav1: Chimera 1080p 10-bitmkl-dnn: IP Batch 1D - f32mkl-dnn: IP Batch All - f32mkl-dnn: IP Batch 1D - u8s8f32mkl-dnn: IP Batch All - u8s8f32mkl-dnn: IP Batch 1D - bf16bf16bf16mkl-dnn: IP Batch All - bf16bf16bf16mkl-dnn: Convolution Batch conv_3d - f32mkl-dnn: Convolution Batch conv_all - f32mkl-dnn: Convolution Batch conv_3d - u8s8f32mkl-dnn: Deconvolution Batch deconv_1d - f32mkl-dnn: Deconvolution Batch deconv_3d - f32mkl-dnn: Convolution Batch conv_alexnet - f32mkl-dnn: Convolution Batch conv_all - u8s8f32mkl-dnn: Deconvolution Batch deconv_all - f32mkl-dnn: Deconvolution Batch deconv_1d - u8s8f32mkl-dnn: Deconvolution Batch deconv_3d - u8s8f32mkl-dnn: Recurrent Neural Network Training - f32mkl-dnn: Convolution Batch conv_3d - bf16bf16bf16mkl-dnn: Convolution Batch conv_alexnet - u8s8f32mkl-dnn: Convolution Batch conv_all - bf16bf16bf16mkl-dnn: Convolution Batch conv_googlenet_v3 - f32mkl-dnn: Deconvolution Batch deconv_1d - bf16bf16bf16mkl-dnn: Deconvolution Batch deconv_3d - bf16bf16bf16mkl-dnn: Convolution Batch conv_alexnet - bf16bf16bf16mkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32mkl-dnn: Deconvolution Batch deconv_all - bf16bf16bf16mkl-dnn: Convolution Batch conv_googlenet_v3 - bf16bf16bf16pgbench: Buffer Test - Normal Load - Read Onlypgbench: Buffer Test - Normal Load - Read Writepgbench: Buffer Test - Single Thread - Read Onlypgbench: Buffer Test - Single Thread - Read Writepgbench: Buffer Test - Heavy Contention - Read Onlypgbench: Buffer Test - Heavy Contention - Read Write2 x Intel Xeon Gold 61389.0340306.5024.7812.0135.8711.762.4911.742.144.584.78154.756.15851.134556.541.741.9486.682515.591120.200.893950.66246.8612.8060.033108.2639.816.397.11567.6125.392756.70152.96573225.584739.8118109.92256.04608561.844592.85OpenBenchmarking.org

LeelaChessZero

Backend: BLAS

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.22.0Backend: BLAS2 x Intel Xeon Gold 61383691215SE +/- 0.47, N = 159.031. (CXX) g++ options: -lpthread -lz

LeelaChessZero

Backend: Random

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.22.0Backend: Random2 x Intel Xeon Gold 61389K18K27K36K45KSE +/- 365.59, N = 340306.501. (CXX) g++ options: -lpthread -lz

libgav1

Video Input: Chimera 1080p

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Chimera 1080p2 x Intel Xeon Gold 6138612182430SE +/- 0.08, N = 324.781. (CXX) g++ options: -O3 -lpthread

libgav1

Video Input: Summer Nature 4K

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Summer Nature 4K2 x Intel Xeon Gold 61383691215SE +/- 0.19, N = 912.011. (CXX) g++ options: -O3 -lpthread

libgav1

Video Input: Summer Nature 1080p

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Summer Nature 1080p2 x Intel Xeon Gold 6138816243240SE +/- 0.20, N = 335.871. (CXX) g++ options: -O3 -lpthread

libgav1

Video Input: Chimera 1080p 10-bit

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Chimera 1080p 10-bit2 x Intel Xeon Gold 61383691215SE +/- 0.03, N = 311.761. (CXX) g++ options: -O3 -lpthread

MKL-DNN DNNL

Harness: IP Batch 1D - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: f322 x Intel Xeon Gold 61380.56031.12061.68092.24122.8015SE +/- 0.01, N = 32.49MIN: 2.191. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: IP Batch All - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: f322 x Intel Xeon Gold 61383691215SE +/- 0.03, N = 311.74MIN: 11.331. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: IP Batch 1D - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: u8s8f322 x Intel Xeon Gold 61380.48150.9631.44451.9262.4075SE +/- 0.01, N = 32.14MIN: 1.981. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: IP Batch All - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: u8s8f322 x Intel Xeon Gold 61381.03052.0613.09154.1225.1525SE +/- 0.02, N = 34.58MIN: 4.281. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: IP Batch 1D - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: bf16bf16bf162 x Intel Xeon Gold 61381.07552.1513.22654.3025.3775SE +/- 0.00, N = 34.78MIN: 4.461. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: IP Batch All - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: bf16bf16bf162 x Intel Xeon Gold 6138306090120150SE +/- 0.41, N = 3154.75MIN: 116.151. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_3d - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: f322 x Intel Xeon Gold 6138246810SE +/- 0.01, N = 36.15MIN: 5.981. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_all - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: f322 x Intel Xeon Gold 61382004006008001000SE +/- 0.32, N = 3851.13MIN: 837.011. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_3d - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: u8s8f322 x Intel Xeon Gold 613810002000300040005000SE +/- 3.70, N = 34556.54MIN: 4539.321. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_1d - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: f322 x Intel Xeon Gold 61380.39150.7831.17451.5661.9575SE +/- 0.00, N = 31.74MIN: 1.671. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_3d - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: f322 x Intel Xeon Gold 61380.43650.8731.30951.7462.1825SE +/- 0.00, N = 31.94MIN: 1.891. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_alexnet - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: f322 x Intel Xeon Gold 613820406080100SE +/- 0.15, N = 386.68MIN: 83.651. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_all - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: u8s8f322 x Intel Xeon Gold 61385001000150020002500SE +/- 10.73, N = 32515.59MIN: 2490.581. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_all - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: f322 x Intel Xeon Gold 61382004006008001000SE +/- 0.70, N = 31120.20MIN: 1109.471. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f322 x Intel Xeon Gold 61380.20030.40060.60090.80121.0015SE +/- 0.01, N = 30.89MIN: 0.851. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f322 x Intel Xeon Gold 61388001600240032004000SE +/- 5.13, N = 33950.66MIN: 3935.281. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Recurrent Neural Network Training - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Recurrent Neural Network Training - Data Type: f322 x Intel Xeon Gold 613850100150200250SE +/- 3.23, N = 4246.86MIN: 232.61. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_3d - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: bf16bf16bf162 x Intel Xeon Gold 61383691215SE +/- 0.01, N = 312.80MIN: 12.611. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: u8s8f322 x Intel Xeon Gold 61381326395265SE +/- 0.09, N = 360.03MIN: 57.371. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_all - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: bf16bf16bf162 x Intel Xeon Gold 61387001400210028003500SE +/- 0.74, N = 33108.26MIN: 3096.571. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: f322 x Intel Xeon Gold 6138918273645SE +/- 0.33, N = 339.81MIN: 37.311. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf162 x Intel Xeon Gold 6138246810SE +/- 0.00, N = 36.39MIN: 6.291. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf162 x Intel Xeon Gold 6138246810SE +/- 0.01, N = 37.11MIN: 7.041. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_alexnet - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: bf16bf16bf162 x Intel Xeon Gold 6138120240360480600SE +/- 1.75, N = 3567.61MIN: 562.421. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f322 x Intel Xeon Gold 6138612182430SE +/- 0.10, N = 325.39MIN: 24.461. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Deconvolution Batch deconv_all - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: bf16bf16bf162 x Intel Xeon Gold 61386001200180024003000SE +/- 1.64, N = 32756.70MIN: 2723.661. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

MKL-DNN DNNL

Harness: Convolution Batch conv_googlenet_v3 - Data Type: bf16bf16bf16

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: bf16bf16bf162 x Intel Xeon Gold 6138306090120150SE +/- 0.69, N = 3152.96MIN: 150.111. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read Only2 x Intel Xeon Gold 6138120K240K360K480K600KSE +/- 3790.66, N = 3573225.581. (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.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read Write2 x Intel Xeon Gold 613810002000300040005000SE +/- 45.05, N = 124739.811. (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.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Single Thread - Mode: Read Only2 x Intel Xeon Gold 61384K8K12K16K20KSE +/- 60.30, N = 318109.921. (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.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Single Thread - Mode: Read Write2 x Intel Xeon Gold 613860120180240300SE +/- 0.48, N = 3256.041. (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.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only2 x Intel Xeon Gold 6138130K260K390K520K650KSE +/- 3362.73, N = 3608561.841. (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.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write2 x Intel Xeon Gold 613810002000300040005000SE +/- 76.41, N = 34592.851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Phoronix Test Suite v10.8.4