oneDNN 2.0 Intel Tiger Lake

Intel Core i7-1165G7 testing with a Dell 0GG9PT (1.0.3 BIOS) and Intel UHD 3GB on Ubuntu 20.10 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 2012203-PTS-ONEDNN2092
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Run 1
December 20 2020
  57 Minutes
Run 2
December 20 2020
  57 Minutes
Run 3
December 20 2020
  56 Minutes
Invert Hiding All Results Option
  57 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

Run 1

Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Run 2

Run 3

Processor: Intel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads), Motherboard: Dell 0GG9PT (1.0.3 BIOS), Chipset: Intel Tiger Lake-LP, Memory: 16GB, Disk: Kioxia KBG40ZNS256G NVMe 256GB, Graphics: Intel UHD 3GB (1300MHz), Audio: Realtek ALC289, Network: Intel Wi-Fi 6 AX201

OS: Ubuntu 20.10, Kernel: 5.10.0-051000rc7daily20201213-generic (x86_64) 20201212, Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1, Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1200

oneDNN 2.0 Intel Tiger LakeOpenBenchmarking.orgPhoronix Test SuiteIntel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads)Dell 0GG9PT (1.0.3 BIOS)Intel Tiger Lake-LP16GBKioxia KBG40ZNS256G NVMe 256GBIntel UHD 3GB (1300MHz)Realtek ALC289Intel Wi-Fi 6 AX201Ubuntu 20.105.10.0-051000rc7daily20201213-generic (x86_64) 20201212GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 Mesa 20.2.11.2.145GCC 10.2.0ext41920x1200ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionOneDNN 2.0 Intel Tiger Lake BenchmarksSystem Logs- --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3 - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Run 1Run 2Run 3Result OverviewPhoronix Test Suite100%103%107%110%oneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNND.B.s - bf16bf16bf16 - CPUIP Shapes 3D - f32 - CPUC.B.S.A - f32 - CPUIP Shapes 3D - u8s8f32 - CPUM.M.B.S.T - u8s8f32 - CPUIP Shapes 3D - bf16bf16bf16 - CPUIP Shapes 1D - u8s8f32 - CPUM.M.B.S.T - f32 - CPUD.B.s - f32 - CPUR.N.N.I - bf16bf16bf16 - CPUR.N.N.I - u8s8f32 - CPUD.B.s - f32 - CPUD.B.s - u8s8f32 - CPUM.M.B.S.T - bf16bf16bf16 - CPUD.B.s - u8s8f32 - CPUR.N.N.I - f32 - CPUR.N.N.T - f32 - CPUC.B.S.A - u8s8f32 - CPUR.N.N.T - bf16bf16bf16 - CPUIP Shapes 1D - bf16bf16bf16 - CPUD.B.s - bf16bf16bf16 - CPUR.N.N.T - u8s8f32 - CPUC.B.S.A - bf16bf16bf16 - CPUIP Shapes 1D - f32 - CPU

oneDNN 2.0 Intel Tiger Lakeonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPURun 1Run 2Run 312.195666.436062.489372.8400825.56117.9205513.497014.099711.354539.647803.008012.742998928.894570.329072.2650.885659.943737.74974537.864.963668891.894532.642.7477112.055312.231416.446132.499322.8530825.76138.2922214.959214.936912.1176510.093983.169322.881299301.854793.379309.3751.965161.726242.82094799.985.351029302.704807.712.9932012.655112.289257.175602.707143.1149626.35078.6347814.111114.924312.1534610.075883.170512.892649350.264799.779315.5752.118961.154142.79314809.375.350079299.604816.403.0055312.7105OpenBenchmarking.org

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 Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPURun 1Run 2Run 33691215SE +/- 0.46, N = 12SE +/- 0.49, N = 12SE +/- 0.47, N = 1212.2012.2312.29MIN: 6.66MIN: 6.4MIN: 5.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620Min: 7.14 / Avg: 12.2 / Max: 12.78Min: 6.83 / Avg: 12.23 / Max: 12.85Min: 7.13 / Avg: 12.29 / Max: 12.931. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPURun 1Run 2Run 3246810SE +/- 0.05431, N = 14SE +/- 0.06535, N = 13SE +/- 0.06630, N = 76.436066.446137.17560MIN: 5.61MIN: 5.6MIN: 5.531. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPURun 1Run 2Run 33691215Min: 5.76 / Avg: 6.44 / Max: 6.58Min: 5.69 / Avg: 6.45 / Max: 6.62Min: 6.79 / Avg: 7.18 / Max: 7.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 30.60911.21821.82732.43643.0455SE +/- 0.03167, N = 12SE +/- 0.03239, N = 12SE +/- 0.02378, N = 122.489372.499322.70714MIN: 1.52MIN: 1.52MIN: 1.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 3246810Min: 2.14 / Avg: 2.49 / Max: 2.53Min: 2.14 / Avg: 2.5 / Max: 2.54Min: 2.45 / Avg: 2.71 / Max: 2.751. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 30.70091.40182.10272.80363.5045SE +/- 0.04400, N = 12SE +/- 0.05466, N = 12SE +/- 0.04263, N = 122.840082.853083.11496MIN: 2.14MIN: 2.14MIN: 2.111. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 3246810Min: 2.36 / Avg: 2.84 / Max: 2.89Min: 2.25 / Avg: 2.85 / Max: 2.92Min: 2.67 / Avg: 3.11 / Max: 3.191. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 3612182430SE +/- 0.58, N = 12SE +/- 0.55, N = 12SE +/- 0.50, N = 1225.5625.7626.35MIN: 18.57MIN: 18.62MIN: 18.71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 3612182430Min: 19.23 / Avg: 25.56 / Max: 26.2Min: 19.72 / Avg: 25.76 / Max: 26.38Min: 20.84 / Avg: 26.35 / Max: 26.891. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 3246810SE +/- 0.11029, N = 12SE +/- 0.29805, N = 12SE +/- 0.13054, N = 127.920558.292228.63478MIN: 4.88MIN: 4.94MIN: 4.721. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 33691215Min: 6.71 / Avg: 7.92 / Max: 8.11Min: 6.69 / Avg: 8.29 / Max: 11.2Min: 7.21 / Avg: 8.63 / Max: 8.891. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620SE +/- 0.19, N = 12SE +/- 0.69, N = 15SE +/- 0.16, N = 1213.5014.9614.11MIN: 8.19MIN: 8.31MIN: 8.281. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620Min: 11.42 / Avg: 13.5 / Max: 13.8Min: 14.23 / Avg: 14.96 / Max: 24.6Min: 12.33 / Avg: 14.11 / Max: 14.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620SE +/- 0.15, N = 4SE +/- 0.07, N = 3SE +/- 0.08, N = 314.1014.9414.92MIN: 12.17MIN: 12.23MIN: 12.231. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620Min: 13.7 / Avg: 14.1 / Max: 14.41Min: 14.79 / Avg: 14.94 / Max: 15.02Min: 14.76 / Avg: 14.92 / Max: 15.011. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPURun 1Run 2Run 33691215SE +/- 0.16, N = 14SE +/- 0.17, N = 15SE +/- 0.18, N = 1411.3512.1212.15MIN: 9.25MIN: 9.29MIN: 9.291. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPURun 1Run 2Run 348121620Min: 9.42 / Avg: 11.35 / Max: 11.66Min: 9.95 / Avg: 12.12 / Max: 12.4Min: 9.94 / Avg: 12.15 / Max: 12.471. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 33691215SE +/- 0.08613, N = 13SE +/- 0.09631, N = 12SE +/- 0.08167, N = 129.6478010.0939810.07588MIN: 7.94MIN: 7.96MIN: 7.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 33691215Min: 8.64 / Avg: 9.65 / Max: 9.83Min: 9.04 / Avg: 10.09 / Max: 10.25Min: 9.18 / Avg: 10.08 / Max: 10.211. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 30.71341.42682.14022.85363.567SE +/- 0.02778, N = 3SE +/- 0.02553, N = 3SE +/- 0.02450, N = 33.008013.169323.17051MIN: 2.49MIN: 2.53MIN: 2.61. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 3246810Min: 2.95 / Avg: 3.01 / Max: 3.04Min: 3.12 / Avg: 3.17 / Max: 3.2Min: 3.12 / Avg: 3.17 / Max: 3.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 30.65081.30161.95242.60323.254SE +/- 0.03844, N = 15SE +/- 0.04127, N = 15SE +/- 0.04298, N = 152.742992.881292.89264MIN: 2.2MIN: 2.2MIN: 2.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 3246810Min: 2.28 / Avg: 2.74 / Max: 2.89Min: 2.35 / Avg: 2.88 / Max: 2.96Min: 2.33 / Avg: 2.89 / Max: 2.961. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPURun 1Run 2Run 32K4K6K8K10KSE +/- 7.25, N = 3SE +/- 5.14, N = 3SE +/- 24.45, N = 38928.899301.859350.26MIN: 8884.62MIN: 9240.8MIN: 9262.281. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPURun 1Run 2Run 316003200480064008000Min: 8921.05 / Avg: 8928.89 / Max: 8943.38Min: 9294.66 / Avg: 9301.85 / Max: 9311.81Min: 9318.15 / Avg: 9350.26 / Max: 9398.251. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPURun 1Run 2Run 310002000300040005000SE +/- 2.42, N = 3SE +/- 12.84, N = 3SE +/- 7.04, N = 34570.324793.374799.77MIN: 4526.96MIN: 4728.25MIN: 4734.031. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPURun 1Run 2Run 38001600240032004000Min: 4565.54 / Avg: 4570.32 / Max: 4573.38Min: 4770.61 / Avg: 4793.37 / Max: 4815.04Min: 4785.86 / Avg: 4799.77 / Max: 4808.631. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 32K4K6K8K10KSE +/- 89.00, N = 3SE +/- 10.72, N = 3SE +/- 4.14, N = 39072.269309.379315.57MIN: 8900.41MIN: 9197.72MIN: 9244.671. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 316003200480064008000Min: 8982.5 / Avg: 9072.26 / Max: 9250.26Min: 9291.56 / Avg: 9309.37 / Max: 9328.61Min: 9307.35 / Avg: 9315.57 / Max: 9320.61. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 31224364860SE +/- 0.98, N = 12SE +/- 0.79, N = 12SE +/- 0.83, N = 1250.8951.9752.12MIN: 36.95MIN: 37.03MIN: 36.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 31020304050Min: 40.06 / Avg: 50.89 / Max: 52.14Min: 43.23 / Avg: 51.97 / Max: 52.83Min: 43.08 / Avg: 52.12 / Max: 54.421. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 31428425670SE +/- 0.67, N = 4SE +/- 0.62, N = 3SE +/- 0.82, N = 359.9461.7361.15MIN: 50.55MIN: 51.66MIN: 48.321. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 31224364860Min: 57.94 / Avg: 59.94 / Max: 60.82Min: 60.49 / Avg: 61.73 / Max: 62.37Min: 59.54 / Avg: 61.15 / Max: 62.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 31020304050SE +/- 0.36, N = 3SE +/- 0.54, N = 15SE +/- 0.55, N = 1537.7542.8242.79MIN: 37.26MIN: 37.09MIN: 37.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 3918273645Min: 37.37 / Avg: 37.75 / Max: 38.47Min: 37.51 / Avg: 42.82 / Max: 43.91Min: 37.53 / Avg: 42.79 / Max: 43.771. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 310002000300040005000SE +/- 1.60, N = 3SE +/- 14.63, N = 3SE +/- 4.81, N = 34537.864799.984809.37MIN: 4499.03MIN: 4723.57MIN: 4750.471. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 38001600240032004000Min: 4535.03 / Avg: 4537.86 / Max: 4540.57Min: 4772.78 / Avg: 4799.98 / Max: 4822.91Min: 4800.21 / Avg: 4809.37 / Max: 4816.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPURun 1Run 2Run 31.2042.4083.6124.8166.02SE +/- 0.00981, N = 3SE +/- 0.02405, N = 3SE +/- 0.02155, N = 34.963665.351025.35007MIN: 3.84MIN: 3.85MIN: 3.861. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPURun 1Run 2Run 3246810Min: 4.95 / Avg: 4.96 / Max: 4.98Min: 5.33 / Avg: 5.35 / Max: 5.4Min: 5.31 / Avg: 5.35 / Max: 5.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 32K4K6K8K10KSE +/- 18.40, N = 3SE +/- 6.77, N = 3SE +/- 8.81, N = 38891.899302.709299.60MIN: 8839.03MIN: 9224.29MIN: 9240.311. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 316003200480064008000Min: 8872.82 / Avg: 8891.89 / Max: 8928.68Min: 9295.91 / Avg: 9302.7 / Max: 9316.23Min: 9282.06 / Avg: 9299.6 / Max: 9309.821. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 310002000300040005000SE +/- 3.28, N = 3SE +/- 7.95, N = 3SE +/- 21.46, N = 34532.644807.714816.40MIN: 4496.04MIN: 4750.99MIN: 4739.611. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 38001600240032004000Min: 4527.36 / Avg: 4532.64 / Max: 4538.65Min: 4795.78 / Avg: 4807.71 / Max: 4822.77Min: 4794.4 / Avg: 4816.4 / Max: 4859.321. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 30.67621.35242.02862.70483.381SE +/- 0.00533, N = 3SE +/- 0.01448, N = 3SE +/- 0.01768, N = 32.747712.993203.00553MIN: 2.05MIN: 2.06MIN: 2.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPURun 1Run 2Run 3246810Min: 2.74 / Avg: 2.75 / Max: 2.76Min: 2.97 / Avg: 2.99 / Max: 3.02Min: 2.98 / Avg: 3.01 / Max: 3.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 33691215SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 312.0612.6612.71MIN: 11.05MIN: 11.29MIN: 11.251. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPURun 1Run 2Run 348121620Min: 12.04 / Avg: 12.06 / Max: 12.08Min: 12.61 / Avg: 12.66 / Max: 12.73Min: 12.65 / Avg: 12.71 / Max: 12.781. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Run 1

Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 20 December 2020 10:07 by user pts.

Run 2

Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 20 December 2020 11:22 by user pts.

Run 3

Processor: Intel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads), Motherboard: Dell 0GG9PT (1.0.3 BIOS), Chipset: Intel Tiger Lake-LP, Memory: 16GB, Disk: Kioxia KBG40ZNS256G NVMe 256GB, Graphics: Intel UHD 3GB (1300MHz), Audio: Realtek ALC289, Network: Intel Wi-Fi 6 AX201

OS: Ubuntu 20.10, Kernel: 5.10.0-051000rc7daily20201213-generic (x86_64) 20201212, Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1, Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1200

Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 20 December 2020 12:35 by user pts.