Intel Xeon Max AMX HBM2e Performance Benchmark

Benchmarks for a future article on Phoronix by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2307017-NE-XEONMAXAM71
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
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

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
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Xeon Max 9480 2P, No HBM, Max AVX512 FP16
July 01 2023
  1 Hour, 35 Minutes
Xeon Max 9480 2P, No HBM
June 30 2023
  2 Hours, 13 Minutes
Xeon Max 9480 2P, HBM Caching
June 29 2023
  2 Hours, 13 Minutes
Xeon Max 9480 2P, HBM Only
June 30 2023
  1 Hour, 35 Minutes
EPYC 9554 2P
June 28 2023
  1 Hour, 36 Minutes
EPYC 9654 2P
June 28 2023
  1 Hour, 34 Minutes
Invert Behavior (Only Show Selected Data)
  1 Hour, 48 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):


Intel Xeon Max AMX HBM2e Performance BenchmarkProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P2 x Intel Xeon Max 9480 @ 3.50GHz (112 Cores / 224 Threads)Supermicro X13DEM v1.10 (1.3 BIOS)Intel Device 1bce512GB7682GB INTEL SSDPF2KX076TZASPEEDVE2282 x Broadcom BCM57508 NetXtreme-E 10Gb/25Gb/40Gb/50Gb/100Gb/200GbUbuntu 23.046.2.0-20-generic (x86_64)GNOME Shell 44.0X Server 1.21.1.7GCC 12.2.0ext41920x1080128GB2 x AMD EPYC 9554 64-Core @ 3.10GHz (128 Cores / 256 Threads)AMD Titanite_4G (RTI1007B BIOS)AMD Device 14a41520GBVGA HDMIBroadcom NetXtreme BCM5720 PCIe2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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-Pa930Z/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Pa930Z/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 -v Processor Details- Xeon Max 9480 2P, No HBM, Max AVX512 FP16: Scaling Governor: intel_cpufreq performance - CPU Microcode: 0x2c0001d1- Xeon Max 9480 2P, No HBM: Scaling Governor: intel_cpufreq performance - CPU Microcode: 0x2c0001d1- Xeon Max 9480 2P, HBM Caching: Scaling Governor: intel_cpufreq performance - CPU Microcode: 0x2c0001d1- Xeon Max 9480 2P, HBM Only: Scaling Governor: intel_cpufreq performance - CPU Microcode: 0x2c0001d1- EPYC 9554 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121- EPYC 9654 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121Python Details- Python 3.11.2Security Details- Xeon Max 9480 2P, No HBM, Max AVX512 FP16: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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 - Xeon Max 9480 2P, No HBM: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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 - Xeon Max 9480 2P, HBM Caching: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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 - Xeon Max 9480 2P, HBM Only: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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 - EPYC 9554 2P: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9654 2P: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Xeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2PLogarithmic Result OverviewPhoronix Test SuiteOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOOpenVINOONNX RuntimeOpenVINOOpenVINOONNX RuntimeONNX RuntimeONNX RuntimeW.P.D.F - CPUF.D.F - CPUF.D.F - CPUP.D.F - CPUP.D.F - CPUF.D.F.I - CPUP.D.F - CPUW.P.D.F.I - CPUP.D.F - CPUW.P.D.F.I - CPUGPT-2 - CPU - StandardW.P.D.F - CPUF.D.F.I - CPUCaffeNet 12-int8 - CPU - StandardCaffeNet 12-int8 - CPU - StandardGPT-2 - CPU - Standard

Intel Xeon Max AMX HBM2e Performance Benchmarkonnx: CaffeNet 12-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: GPT-2 - CPU - Standardopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Face Detection FP16 - CPUopenvino: Face Detection FP16 - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P1.83797543.8116.41692157.758262.84106.272445.4911.311398.3219.832449.2911.2610.4710680.9556.601977.081.86415537.8216.50889155.155380.60293.49857.3232.52242.64115.20824.9333.793.4531482.786.6516532.401.78296562.2184.70849218.692333.51334.89834.9733.40232.01120.40819.7633.973.4730938.506.4416941.991.68307596.9084.05237246.700302.45369.64706.8339.47159.13175.74719.4938.773.0735643.795.9218452.022.01949495.2407.64735132.721205.71155.28832.0238.24389.5382.04825.5238.538.0915659.323.978039.572.30611435.1949.68904103.184250.39191.411092.9743.64471.82101.561098.9643.409.9519170.614.869843.03OpenBenchmarking.org

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P0.51891.03781.55672.07562.5945SE +/- 0.01109, N = 3SE +/- 0.02958, N = 15SE +/- 0.02850, N = 15SE +/- 0.03485, N = 15SE +/- 0.02472, N = 4SE +/- 0.03734, N = 151.837971.864151.782961.683072.019492.306111. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P130260390520650SE +/- 3.29, N = 3SE +/- 7.97, N = 15SE +/- 8.12, N = 15SE +/- 10.95, N = 15SE +/- 5.94, N = 4SE +/- 7.41, N = 15543.81537.82562.22596.91495.24435.191. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: StandardXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P3691215SE +/- 0.19970, N = 15SE +/- 0.18098, N = 15SE +/- 0.23765, N = 15SE +/- 0.05143, N = 3SE +/- 0.25707, N = 15SE +/- 0.00311, N = 36.416926.508894.708494.052377.647359.689041. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: StandardXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P50100150200250SE +/- 4.54, N = 15SE +/- 4.09, N = 15SE +/- 9.23, N = 15SE +/- 3.11, N = 3SE +/- 4.25, N = 15SE +/- 0.03, N = 3157.76155.16218.69246.70132.72103.181. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P80160240320400SE +/- 0.33, N = 3SE +/- 0.34, N = 3SE +/- 0.44, N = 3SE +/- 0.04, N = 3SE +/- 0.22, N = 3SE +/- 0.04, N = 3262.84380.60333.51302.45205.71250.39MIN: 192.97 / MAX: 452.59MIN: 286.27 / MAX: 602.06MIN: 285.32 / MAX: 516.28MIN: 217.26 / MAX: 450.48MIN: 202.38 / MAX: 252.14MIN: 222.36 / MAX: 303.721. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P80160240320400SE +/- 0.10, N = 3SE +/- 0.25, N = 3SE +/- 0.45, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 3SE +/- 0.06, N = 3106.27293.49334.89369.64155.28191.411. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P5001000150020002500SE +/- 9.42, N = 3SE +/- 10.44, N = 3SE +/- 6.17, N = 3SE +/- 1.42, N = 3SE +/- 5.30, N = 3SE +/- 5.11, N = 32445.49857.32834.97706.83832.021092.97MIN: 1611.66 / MAX: 3114.74MIN: 481.73 / MAX: 1422.59MIN: 558.93 / MAX: 2258.39MIN: 455.92 / MAX: 1885.53MIN: 743.95 / MAX: 1258.17MIN: 825.28 / MAX: 1814.821. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P1020304050SE +/- 0.04, N = 3SE +/- 0.42, N = 3SE +/- 0.24, N = 3SE +/- 0.08, N = 3SE +/- 0.23, N = 3SE +/- 0.21, N = 311.3132.5233.4039.4738.2443.641. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P30060090012001500SE +/- 8.00, N = 3SE +/- 0.61, N = 3SE +/- 0.80, N = 3SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.93, N = 31398.32242.64232.01159.13389.53471.82MIN: 943.82 / MAX: 1957.07MIN: 380.82 / MAX: 432.57MIN: 431.51 / MAX: 569.31. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P4080120160200SE +/- 0.11, N = 3SE +/- 0.29, N = 3SE +/- 0.41, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.25, N = 319.83115.20120.40175.7482.04101.561. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P5001000150020002500SE +/- 11.98, N = 3SE +/- 6.05, N = 3SE +/- 7.20, N = 3SE +/- 7.70, N = 3SE +/- 1.07, N = 3SE +/- 5.33, N = 32449.29824.93819.76719.49825.521098.96MIN: 1558.62 / MAX: 3306.73MIN: 445.3 / MAX: 1462.45MIN: 516.88 / MAX: 2598.39MIN: 473.54 / MAX: 2264.09MIN: 747.05 / MAX: 1238.39MIN: 825.62 / MAX: 1835.761. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P1020304050SE +/- 0.07, N = 3SE +/- 0.25, N = 3SE +/- 0.31, N = 3SE +/- 0.42, N = 3SE +/- 0.04, N = 3SE +/- 0.20, N = 311.2633.7933.9738.7738.5343.401. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P3691215SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 310.473.453.473.078.099.95MIN: 7.24 / MAX: 56.73MIN: 2.47 / MAX: 46.27MIN: 2.48 / MAX: 56.57MIN: 2.48 / MAX: 68.92MIN: 7.89 / MAX: 45.39MIN: 8.36 / MAX: 40.391. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P8K16K24K32K40KSE +/- 12.56, N = 3SE +/- 37.83, N = 3SE +/- 38.32, N = 3SE +/- 37.65, N = 3SE +/- 3.60, N = 3SE +/- 4.81, N = 310680.9531482.7830938.5035643.7915659.3219170.611. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P1326395265SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 356.606.656.445.923.974.86MIN: 44.94 / MAX: 112.121. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P4K8K12K16K20KSE +/- 2.85, N = 3SE +/- 65.25, N = 3SE +/- 13.63, N = 3SE +/- 33.75, N = 3SE +/- 1.22, N = 3SE +/- 2.32, N = 31977.0816532.4016941.9918452.028039.579843.031. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

System Power Consumption Monitor

OpenBenchmarking.orgWattsSystem Power Consumption MonitorPhoronix Test Suite System MonitoringXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM Only160320480640800Min: 276 / Avg: 733.34 / Max: 937Min: 251 / Avg: 735.78 / Max: 904Min: 260 / Avg: 751.74 / Max: 907Min: 250 / Avg: 752.26 / Max: 929

CPU Peak Freq (Highest CPU Core Frequency) Monitor

OpenBenchmarking.orgMegahertzCPU Peak Freq (Highest CPU Core Frequency) MonitorPhoronix Test Suite System MonitoringXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM Only6001200180024003000Min: 1800 / Avg: 3336.55 / Max: 3518Min: 1000 / Avg: 3056.85 / Max: 3576Min: 1100 / Avg: 3207.85 / Max: 3577Min: 1566 / Avg: 3308.69 / Max: 3551

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringXeon Max 9480 2P, No HBM, Max AVX512 FP16Xeon Max 9480 2P, No HBMXeon Max 9480 2P, HBM CachingXeon Max 9480 2P, HBM OnlyEPYC 9554 2PEPYC 9654 2P2004006008001000Min: 166.66 / Avg: 528.06 / Max: 636.62Min: 93.37 / Avg: 522.99 / Max: 664.03Min: 87.33 / Avg: 548.37 / Max: 899.87Min: 129.15 / Avg: 563.95 / Max: 1026.74Min: 34.39 / Avg: 413.82 / Max: 710.63Min: 31.88 / Avg: 448.81 / Max: 725.51

19 Results Shown

ONNX Runtime:
  CaffeNet 12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  GPT-2 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Face Detection FP16-INT8 - CPU:
    ms
    FPS
  Person Detection FP16 - CPU:
    ms
    FPS
  Face Detection FP16 - CPU:
    ms
    FPS
  Person Detection FP32 - CPU:
    ms
    FPS
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
  Weld Porosity Detection FP16 - CPU:
    ms
    FPS
System Power Consumption Monitor:
  Phoronix Test Suite System Monitoring:
    Watts
    Megahertz
    Watts