Benchmarks for a future article. Intel Core i9-11900K testing with a ASUS ROG MAXIMUS XIII HERO (1402 BIOS) and ASUS Intel RKL GT1 31GB on Ubuntu 22.10 via the Phoronix Test Suite.
i9-11900K: AVX-512 On Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
i9-11900K: AVX-512 Off Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -mno-avx512f" CFLAGS="-O3 -march=native -mno-avx512f"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
i9-11900K: AVX-512 On 512 Processor: Intel Core i9-11900K @ 5.10GHz (8 Cores / 16 Threads), Motherboard: ASUS ROG MAXIMUS XIII HERO (1402 BIOS), Chipset: Intel Tiger Lake-H, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 32GB Flash Drive, Graphics: ASUS Intel RKL GT1 31GB (1300MHz), Audio: Intel Tiger Lake-H HD Audio, Monitor: ASUS MG28U, Network: 2 x Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.10, Kernel: 5.19.0-21-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -mprefer-vector-width=512" CFLAGS="-O3 -march=native -mprefer-vector-width=512"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 28.26 34.51 28.22 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4651 / Avg: 4793.56 / Max: 5300 Min: 4700 / Avg: 4850.7 / Max: 5301 Min: 4669 / Avg: 4825.21 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.52 / Avg: 194.6 / Max: 244.04 Min: 12.18 / Avg: 155.44 / Max: 200.14 Min: 12.22 / Avg: 188.58 / Max: 244.57 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 79.78 / Max: 91 Min: 35 / Avg: 69.17 / Max: 81 Min: 38 / Avg: 77.04 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 7 14 21 28 35 Min: 28.17 / Avg: 28.26 / Max: 28.36 Min: 34.47 / Avg: 34.51 / Max: 34.55 Min: 28.06 / Avg: 28.22 / Max: 28.34
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 30 60 90 120 150 SE +/- 0.31, N = 3 SE +/- 0.07, N = 3 SE +/- 0.54, N = 3 97.85 112.60 98.03 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4550 / Avg: 4788.85 / Max: 5300 Min: 4700 / Avg: 4864.38 / Max: 5300 Min: 3500 / Avg: 4771.09 / Max: 5310 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.48 / Avg: 190.62 / Max: 260.18 Min: 12.38 / Avg: 167.33 / Max: 219.34 Min: 12.38 / Avg: 190.62 / Max: 262.43 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 40 / Avg: 78.33 / Max: 92 Min: 38 / Avg: 72.32 / Max: 85 Min: 39 / Avg: 78.25 / Max: 94 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 97.29 / Avg: 97.85 / Max: 98.35 Min: 112.47 / Avg: 112.6 / Max: 112.7 Min: 97.26 / Avg: 98.03 / Max: 99.07
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 14.54 17.66 14.54 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4769.5 / Max: 5309 Min: 4700 / Avg: 4829.28 / Max: 5301 Min: 4700 / Avg: 4800.61 / Max: 5310 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.29 / Avg: 180.67 / Max: 210.72 Min: 12.39 / Avg: 153.63 / Max: 193.33 Min: 12.32 / Avg: 171.93 / Max: 211.04 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 15 30 45 60 75 Min: 40 / Avg: 72.9 / Max: 79 Min: 38 / Avg: 66.44 / Max: 73 Min: 38 / Avg: 69.89 / Max: 79 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 4 8 12 16 20 Min: 14.52 / Avg: 14.54 / Max: 14.56 Min: 17.65 / Avg: 17.66 / Max: 17.68 Min: 14.53 / Avg: 14.54 / Max: 14.55
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 13 26 39 52 65 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 49.57 57.88 49.36 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4777.02 / Max: 5300 Min: 4700 / Avg: 4802.88 / Max: 5301 Min: 4700 / Avg: 4777.6 / Max: 5301 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.55 / Avg: 190.88 / Max: 225.22 Min: 12.42 / Avg: 169 / Max: 204.27 Min: 12.41 / Avg: 191.37 / Max: 225.56 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 16 32 48 64 80 Min: 40 / Avg: 73.39 / Max: 81 Min: 39 / Avg: 69.87 / Max: 78 Min: 39 / Avg: 73.25 / Max: 82 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 11 22 33 44 55 Min: 49.43 / Avg: 49.57 / Max: 49.67 Min: 57.83 / Avg: 57.88 / Max: 57.93 Min: 49.3 / Avg: 49.36 / Max: 49.41
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 3 6 9 12 15 SE +/- 0.0274, N = 3 SE +/- 0.0062, N = 3 SE +/- 0.0069, N = 3 7.5173 11.6244 7.4707 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3556 / Avg: 4766.8 / Max: 5302 Min: 4700 / Avg: 4780.8 / Max: 5312 Min: 4631 / Avg: 4780.28 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.32 / Avg: 201.37 / Max: 246.43 Min: 12.35 / Avg: 183.63 / Max: 214.9 Min: 12.3 / Avg: 209.51 / Max: 246.88 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 38 / Avg: 79.49 / Max: 92 Min: 39 / Avg: 74.18 / Max: 83 Min: 39 / Avg: 81.82 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 3 6 9 12 15 Min: 7.48 / Avg: 7.52 / Max: 7.57 Min: 11.62 / Avg: 11.62 / Max: 11.64 Min: 7.46 / Avg: 7.47 / Max: 7.48
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 9 18 27 36 45 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 25.73 39.35 25.49 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3500 / Avg: 4648.5 / Max: 5302 Min: 4700 / Avg: 4768.47 / Max: 5302 Min: 3500 / Avg: 4682.19 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.48 / Avg: 211.14 / Max: 256.86 Min: 12.46 / Avg: 198.19 / Max: 227.43 Min: 12.31 / Avg: 211.91 / Max: 258.7 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 40 / Avg: 82.42 / Max: 94 Min: 40 / Avg: 78.22 / Max: 86 Min: 41 / Avg: 82.64 / Max: 96 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 8 16 24 32 40 Min: 25.5 / Avg: 25.73 / Max: 26.01 Min: 39.24 / Avg: 39.34 / Max: 39.41 Min: 25.35 / Avg: 25.49 / Max: 25.71
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 30 60 90 120 150 SE +/- 1.00, N = 3 SE +/- 0.12, N = 3 SE +/- 0.37, N = 3 112.68 128.05 111.22 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3500 / Avg: 4747.88 / Max: 5310 Min: 4700 / Avg: 4847.56 / Max: 5310 Min: 4575 / Avg: 4802.05 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.46 / Avg: 192.59 / Max: 258.71 Min: 12.38 / Avg: 167.69 / Max: 217.29 Min: 12.38 / Avg: 193.85 / Max: 256.16 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 78.7 / Max: 94 Min: 38 / Avg: 71.71 / Max: 85 Min: 39 / Avg: 79.14 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 111.12 / Avg: 112.68 / Max: 114.54 Min: 127.87 / Avg: 128.05 / Max: 128.27 Min: 110.49 / Avg: 111.22 / Max: 111.7
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 110 220 330 440 550 SE +/- 3.25, N = 3 SE +/- 0.40, N = 3 SE +/- 0.68, N = 3 446.88 492.38 440.10 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3500 / Avg: 4711.32 / Max: 5304 Min: 4700 / Avg: 4862.17 / Max: 5301 Min: 4389 / Avg: 4809.78 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.5 / Avg: 186.42 / Max: 260.13 Min: 12.43 / Avg: 168.83 / Max: 221.42 Min: 12.39 / Avg: 189.75 / Max: 259.61 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 41 / Avg: 76.64 / Max: 95 Min: 39 / Avg: 72.83 / Max: 87 Min: 40 / Avg: 78.69 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 90 180 270 360 450 Min: 442.44 / Avg: 446.88 / Max: 453.21 Min: 491.91 / Avg: 492.38 / Max: 493.18 Min: 438.75 / Avg: 440.1 / Max: 440.88
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 9 18 27 36 45 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 22.89 39.14 23.01 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4810.87 / Max: 5300 Min: 4700 / Avg: 4884.74 / Max: 5310 Min: 4700 / Avg: 4865.03 / Max: 5312 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.24 / Avg: 188.78 / Max: 249.74 Min: 12.29 / Avg: 152.74 / Max: 206.29 Min: 12.27 / Avg: 178.55 / Max: 250.04 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 75.07 / Max: 90 Min: 38 / Avg: 64.6 / Max: 73 Min: 38 / Avg: 73.65 / Max: 89 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 8 16 24 32 40 Min: 22.71 / Avg: 22.89 / Max: 23.07 Min: 39.03 / Avg: 39.14 / Max: 39.3 Min: 22.83 / Avg: 23.01 / Max: 23.17
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.85, N = 3 SE +/- 0.07, N = 3 109.28 150.77 108.49 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4803.42 / Max: 5300 Min: 4700 / Avg: 4899.77 / Max: 5314 Min: 4700 / Avg: 4813.34 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.58 / Avg: 186.84 / Max: 241.51 Min: 12.57 / Avg: 151.3 / Max: 208.63 Min: 12.39 / Avg: 185.33 / Max: 241.17 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 40 / Avg: 74.99 / Max: 88 Min: 39 / Avg: 65.01 / Max: 75 Min: 37 / Avg: 73.66 / Max: 88 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 30 60 90 120 150 Min: 109.12 / Avg: 109.28 / Max: 109.44 Min: 149.8 / Avg: 150.77 / Max: 152.45 Min: 108.38 / Avg: 108.49 / Max: 108.61
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 20.13 23.04 19.99 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4785.87 / Max: 5309 Min: 4700 / Avg: 4792.9 / Max: 5300 Min: 4700 / Avg: 4796.86 / Max: 5302 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.29 / Avg: 195.66 / Max: 239.75 Min: 12.46 / Avg: 178.55 / Max: 211.42 Min: 12.46 / Avg: 200.45 / Max: 240.26 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 80.23 / Max: 93 Min: 38 / Avg: 73.22 / Max: 82 Min: 39 / Avg: 80.25 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 5 10 15 20 25 Min: 20.04 / Avg: 20.13 / Max: 20.24 Min: 23.03 / Avg: 23.04 / Max: 23.04 Min: 19.97 / Avg: 19.99 / Max: 20.03
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 SE +/- 0.23, N = 3 SE +/- 0.19, N = 3 SE +/- 0.56, N = 3 78.78 92.51 77.87 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3578 / Avg: 4752.98 / Max: 5300 Min: 4700 / Avg: 4785.41 / Max: 5300 Min: 4700 / Avg: 4783.64 / Max: 5301 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 Min: 12.65 / Avg: 200.79 / Max: 239.52 Min: 12.46 / Avg: 181.5 / Max: 215.22 Min: 12.46 / Avg: 204.1 / Max: 244.9 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 40 / Avg: 80.82 / Max: 93 Min: 39 / Avg: 74.07 / Max: 85 Min: 39 / Avg: 80.84 / Max: 95 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 78.35 / Avg: 78.78 / Max: 79.11 Min: 92.13 / Avg: 92.51 / Max: 92.7 Min: 76.76 / Avg: 77.87 / Max: 78.44
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 30 60 90 120 150 SE +/- 0.58, N = 3 SE +/- 0.07, N = 3 SE +/- 0.33, N = 3 112.19 127.65 110.96 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3500 / Avg: 4768.23 / Max: 5302 Min: 4700 / Avg: 4854.95 / Max: 5363 Min: 4587 / Avg: 4805.51 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.46 / Avg: 191.4 / Max: 257.39 Min: 12.27 / Avg: 168.07 / Max: 217.66 Min: 12.35 / Avg: 194.31 / Max: 256.42 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 38 / Avg: 79.14 / Max: 93 Min: 38 / Avg: 71.94 / Max: 85 Min: 39 / Avg: 79.22 / Max: 92 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 111.06 / Avg: 112.19 / Max: 113 Min: 127.52 / Avg: 127.65 / Max: 127.74 Min: 110.31 / Avg: 110.96 / Max: 111.33
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 110 220 330 440 550 SE +/- 1.90, N = 3 SE +/- 0.41, N = 3 SE +/- 1.29, N = 3 446.49 491.12 440.56 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 3500 / Avg: 4712.83 / Max: 5302 Min: 4700 / Avg: 4864.76 / Max: 5300 Min: 3500 / Avg: 4792.63 / Max: 5301 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.69 / Avg: 185.26 / Max: 260.57 Min: 12.4 / Avg: 167.72 / Max: 220.91 Min: 12.54 / Avg: 189.06 / Max: 260.04 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.1 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 77.07 / Max: 96 Min: 39 / Avg: 72.53 / Max: 86 Min: 40 / Avg: 78.84 / Max: 95 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream AVX-512 On AVX-512 Off AVX-512 On 512 90 180 270 360 450 Min: 442.69 / Avg: 446.49 / Max: 448.42 Min: 490.53 / Avg: 491.12 / Max: 491.91 Min: 439.04 / Avg: 440.56 / Max: 443.13
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time AVX-512 On AVX-512 Off AVX-512 On 512 11 22 33 44 55 45.49 46.53 45.33 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU
i9-11900K: AVX-512 Off: The test run did not produce a result.
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Triple SHA-256, Onecoin AVX-512 On AVX-512 Off AVX-512 On 512 50K 100K 150K 200K 250K SE +/- 1895.81, N = 3 SE +/- 627.84, N = 3 SE +/- 1461.35, N = 3 231700 106137 230687 -mno-avx512f 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.18 Algorithm: Triple SHA-256, Onecoin AVX-512 On AVX-512 Off AVX-512 On 512 300 600 900 1200 1500 1366.28 820.32 1361.66 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Cpuminer-Opt 3.18 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4759.14 / Max: 5300 Min: 4700 / Avg: 4834.78 / Max: 5300 Min: 4700 / Avg: 4753.44 / Max: 5300 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Cpuminer-Opt 3.18 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.5 / Avg: 169.58 / Max: 237.02 Min: 12.4 / Avg: 129.38 / Max: 169.44 Min: 12.41 / Avg: 169.42 / Max: 278.89 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Cpuminer-Opt 3.18 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 40 / Avg: 74.89 / Max: 92 Min: 39 / Avg: 64.76 / Max: 73 Min: 42 / Avg: 75.38 / Max: 94 Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Triple SHA-256, Onecoin AVX-512 On AVX-512 Off AVX-512 On 512 40K 80K 120K 160K 200K Min: 227960 / Avg: 231700 / Max: 234110 Min: 104890 / Avg: 106136.67 / Max: 106890 Min: 228060 / Avg: 230686.67 / Max: 233110 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Quad SHA-256, Pyrite AVX-512 On AVX-512 Off AVX-512 On 512 40K 80K 120K 160K 200K SE +/- 280.42, N = 3 SE +/- 42.56, N = 3 SE +/- 346.55, N = 3 167500 64663 164790 -mno-avx512f 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.18 Algorithm: Quad SHA-256, Pyrite AVX-512 On AVX-512 Off AVX-512 On 512 200 400 600 800 1000 953.89 460.38 930.90 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Cpuminer-Opt 3.18 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4750.65 / Max: 5300 Min: 4700 / Avg: 4831.54 / Max: 5312 Min: 4700 / Avg: 4753.87 / Max: 5290 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Cpuminer-Opt 3.18 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.43 / Avg: 175.6 / Max: 252.11 Min: 12.24 / Avg: 140.46 / Max: 186.51 Min: 12.25 / Avg: 177.02 / Max: 245.52 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Cpuminer-Opt 3.18 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 38 / Avg: 73.74 / Max: 92 Min: 36 / Avg: 65.33 / Max: 73 Min: 39 / Avg: 75.84 / Max: 93 Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Quad SHA-256, Pyrite AVX-512 On AVX-512 Off AVX-512 On 512 30K 60K 90K 120K 150K Min: 167030 / Avg: 167500 / Max: 168000 Min: 64580 / Avg: 64663.33 / Max: 64720 Min: 164180 / Avg: 164790 / Max: 165380 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: LBC, LBRY Credits AVX-512 On AVX-512 Off AVX-512 On 512 17K 34K 51K 68K 85K SE +/- 84.13, N = 3 SE +/- 20.00, N = 3 SE +/- 116.81, N = 3 77083 27270 76393 -mno-avx512f 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.18 Algorithm: LBC, LBRY Credits AVX-512 On AVX-512 Off AVX-512 On 512 90 180 270 360 450 397.33 165.49 393.49 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Cpuminer-Opt 3.18 CPU Peak Freq (Highest CPU Core Frequency) Monitor AVX-512 On AVX-512 Off AVX-512 On 512 900 1800 2700 3600 4500 Min: 4700 / Avg: 4754.62 / Max: 5300 Min: 4700 / Avg: 4769.7 / Max: 5300 Min: 4700 / Avg: 4753.32 / Max: 5289 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Cpuminer-Opt 3.18 CPU Power Consumption Monitor AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 12.52 / Avg: 194 / Max: 268.78 Min: 12.49 / Avg: 164.79 / Max: 203.1 Min: 12.24 / Avg: 194.14 / Max: 277.39 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Cpuminer-Opt 3.18 CPU Temperature Monitor AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 39 / Avg: 77.28 / Max: 95 Min: 37 / Avg: 68.15 / Max: 76 Min: 40 / Avg: 77.84 / Max: 95 Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: LBC, LBRY Credits AVX-512 On AVX-512 Off AVX-512 On 512 13K 26K 39K 52K 65K Min: 76920 / Avg: 77083.33 / Max: 77200 Min: 27230 / Avg: 27270 / Max: 27290 Min: 76160 / Avg: 76393.33 / Max: 76520 1. (CXX) g++ options: -O3 -march=native -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.
Device: CPU - Batch Size: 512 - Model: VGG-16
i9-11900K: AVX-512 On: The test quit with a non-zero exit status.
i9-11900K: AVX-512 Off: The test quit with a non-zero exit status. E: Fatal Python error: Aborted
i9-11900K: AVX-512 On 512: The test quit with a non-zero exit status. E: Fatal Python error: Segmentation fault
Device: CPU - Batch Size: 512 - Model: ResNet-50
i9-11900K: AVX-512 On: The test quit with a non-zero exit status.
i9-11900K: AVX-512 Off: The test quit with a non-zero exit status.
i9-11900K: AVX-512 On 512: The test quit with a non-zero exit status.
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.
Meta Performance Per Watts OpenBenchmarking.org Performance Per Watts, More Is Better Meta Performance Per Watts Performance Per Watts AVX-512 On AVX-512 Off AVX-512 On 512 40 80 120 160 200 170.75 133.30 170.21
CPU Peak Freq (Highest CPU Core Frequency) Monitor OpenBenchmarking.org Megahertz CPU Peak Freq (Highest CPU Core Frequency) Monitor Phoronix Test Suite System Monitoring AVX-512 On AVX-512 Off AVX-512 On 512 1000 2000 3000 4000 5000 Min: 3350 / Avg: 4733.96 / Max: 5541 Min: 3500 / Avg: 4781.84 / Max: 5621 Min: 2700 / Avg: 4722.22 / Max: 5323
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring AVX-512 On AVX-512 Off AVX-512 On 512 50 100 150 200 250 Min: 6.37 / Avg: 188.45 / Max: 283.94 Min: 6.36 / Avg: 173.58 / Max: 251.93 Min: 6.3 / Avg: 192.29 / Max: 280.23
CPU Temperature Monitor OpenBenchmarking.org Celsius CPU Temperature Monitor Phoronix Test Suite System Monitoring AVX-512 On AVX-512 Off AVX-512 On 512 20 40 60 80 100 Min: 29 / Avg: 78.17 / Max: 100 Min: 27 / Avg: 72.98 / Max: 95 Min: 31 / Avg: 79.47 / Max: 100
i9-11900K: AVX-512 On Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 18 October 2022 10:35 by user phoronix.
i9-11900K: AVX-512 Off Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -mno-avx512f" CFLAGS="-O3 -march=native -mno-avx512f"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 19 October 2022 09:56 by user phoronix.
i9-11900K: AVX-512 On 512 Processor: Intel Core i9-11900K @ 5.10GHz (8 Cores / 16 Threads), Motherboard: ASUS ROG MAXIMUS XIII HERO (1402 BIOS), Chipset: Intel Tiger Lake-H, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 32GB Flash Drive, Graphics: ASUS Intel RKL GT1 31GB (1300MHz), Audio: Intel Tiger Lake-H HD Audio, Monitor: ASUS MG28U, Network: 2 x Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.10, Kernel: 5.19.0-21-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -mprefer-vector-width=512" CFLAGS="-O3 -march=native -mprefer-vector-width=512"Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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 -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x54 - Thermald 2.5.1Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 20 October 2022 10:52 by user phoronix.