5218-oneDNN 2 x Intel Xeon Gold 5218 testing with a Intel S2600WFT (SE5C620.86B.02.01.0008.031920191559 BIOS) and ASPEED on CentOS Linux 7 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2009213-NE-5218ONEDN12 2 x Intel Xeon Gold 5218 Processor: 2 x Intel Xeon Gold 5218 @ 3.90GHz (32 Cores / 64 Threads), Motherboard: Intel S2600WFT (SE5C620.86B.02.01.0008.031920191559 BIOS), Memory: 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB DDR4-2666MT/s, Disk: 7 x 1600GB INTEL SSDPE2KE016T8, Graphics: ASPEED, Network: 2 x Mellanox MT27710
OS: CentOS Linux 7, Kernel: 3.10.0-1127.el7.x86_64 (x86_64), Compiler: GCC 4.8.5 20150623, File-System: xfs, Screen Resolution: 1024x768
Compiler Notes: --build=x86_64-redhat-linux --disable-libgcj --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,objc,obj-c++,java,fortran,ada,go,lto --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-linker-hash-style=gnu --with-tune=genericProcessor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002cSecurity Notes: SELinux + itlb_multihit: vulnerable + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of Load fences usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full retpoline IBPB + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.4654 0.9308 1.3962 1.8616 2.327 SE +/- 0.00619, N = 3 2.06852 MIN: 1.91 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 7 14 21 28 35 SE +/- 0.05, N = 3 29.46 MIN: 27.92 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.1523 0.3046 0.4569 0.6092 0.7615 SE +/- 0.003884, N = 3 0.676909 MIN: 0.61 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 2 4 6 8 10 SE +/- 0.00984, N = 3 7.95006 MIN: 7.56 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 1.0618 2.1236 3.1854 4.2472 5.309 SE +/- 0.00336, N = 3 4.71908 MIN: 4.51 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 14 28 42 56 70 SE +/- 0.04, N = 3 63.61 MIN: 62.64 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 1.256 2.512 3.768 5.024 6.28 SE +/- 0.03121, N = 3 5.58214 MIN: 4.8 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.5584 1.1168 1.6752 2.2336 2.792 SE +/- 0.00251, N = 3 2.48178 MIN: 2.34 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.6791 1.3582 2.0373 2.7164 3.3955 SE +/- 0.01722, N = 3 3.01832 MIN: 2.94 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 1.1832 2.3664 3.5496 4.7328 5.916 SE +/- 0.01078, N = 3 5.25871 MIN: 5.11 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.1573 0.3146 0.4719 0.6292 0.7865 SE +/- 0.001093, N = 3 0.699011 MIN: 0.63 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.1729 0.3458 0.5187 0.6916 0.8645 SE +/- 0.002329, N = 3 0.768639 MIN: 0.74 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 60 120 180 240 300 SE +/- 1.50, N = 3 286.35 MIN: 274.77 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 20 40 60 80 100 SE +/- 0.10, N = 3 74.85 MIN: 71.6 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 2 4 6 8 10 SE +/- 0.00738, N = 3 8.50077 MIN: 8.36 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 3 6 9 12 15 SE +/- 0.00637, N = 3 9.35296 MIN: 9.02 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 3 6 9 12 15 SE +/- 0.01, N = 3 11.15 MIN: 10.99 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.1481 0.2962 0.4443 0.5924 0.7405 SE +/- 0.001104, N = 3 0.658310 MIN: 0.61 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 2 x Intel Xeon Gold 5218 0.0927 0.1854 0.2781 0.3708 0.4635 SE +/- 0.002937, N = 3 0.411840 MIN: 0.35 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU 2 x Intel Xeon Gold 5218 0.4282 0.8564 1.2846 1.7128 2.141 SE +/- 0.00146, N = 3 1.90325 MIN: 1.83 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -rdynamic -lpthread -ldl
2 x Intel Xeon Gold 5218 Processor: 2 x Intel Xeon Gold 5218 @ 3.90GHz (32 Cores / 64 Threads), Motherboard: Intel S2600WFT (SE5C620.86B.02.01.0008.031920191559 BIOS), Memory: 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB + 16384 MB + 129408 MB DDR4-2666MT/s, Disk: 7 x 1600GB INTEL SSDPE2KE016T8, Graphics: ASPEED, Network: 2 x Mellanox MT27710
OS: CentOS Linux 7, Kernel: 3.10.0-1127.el7.x86_64 (x86_64), Compiler: GCC 4.8.5 20150623, File-System: xfs, Screen Resolution: 1024x768
Compiler Notes: --build=x86_64-redhat-linux --disable-libgcj --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,objc,obj-c++,java,fortran,ada,go,lto --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-linker-hash-style=gnu --with-tune=genericProcessor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002cSecurity Notes: SELinux + itlb_multihit: vulnerable + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of Load fences usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full retpoline IBPB + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable
Testing initiated at 21 September 2020 00:26 by user .