Intel 5th Gen Xeon AVX-512 Comparison

2 x INTEL XEON PLATINUM 8592 AVX-512 Emerald Rapids vs. Sapphire Rapids by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2312146-NE-INTELAVX523
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

Limit displaying results to tests within:

CPU Massive 2 Tests
Creator Workloads 6 Tests
HPC - High Performance Computing 4 Tests
Machine Learning 3 Tests
Multi-Core 6 Tests
Intel oneAPI 6 Tests
Python Tests 2 Tests
Raytracing 2 Tests
Renderers 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
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
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Comparison
Transpose Comparison

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Sapphire Rapids: AVX-512 On
December 02 2023
  3 Hours, 51 Minutes
Sapphire Rapids: AVX-512 Off
December 03 2023
  4 Hours, 20 Minutes
Emerald Rapids: AVX-512 On
December 11 2023
  4 Hours, 22 Minutes
Emerald Rapids: AVX-512 Off
December 11 2023
  4 Hours, 57 Minutes
Invert Hiding All Results Option
  4 Hours, 23 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):


ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelCompilerFile-SystemScreen ResolutionSapphire RapidsEmerald Rapids AVX-512 On AVX-512 Off AVX-512 On AVX-512 Off2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads)Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS)Intel Device 1bce1008GB3201GB Micron_7450_MTFDKCB3T2TFSASPEED2 x Intel X710 for 10GBASE-TUbuntu 23.106.5.0-13-generic (x86_64)GCC 13.2.0ext41920x10802 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads)Quanta Cloud S6Q-MB-MPS (3B05.TEL4P1 BIOS)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/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-build-config=bootstrap-lto-lean --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- Sapphire Rapids: AVX-512 On: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004d0- Sapphire Rapids: AVX-512 Off: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004d0- Emerald Rapids: AVX-512 On: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000161- Emerald Rapids: AVX-512 Off: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000161Python Details- Python 3.11.6Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: 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 / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

minibude: OpenMP - BM1minibude: OpenMP - BM1minibude: OpenMP - BM2minibude: OpenMP - BM2embree: Pathtracer ISPC - Crownembree: Pathtracer ISPC - Asian Dragonembree: Pathtracer ISPC - Asian Dragon Objopenvkl: vklBenchmarkCPU ISPCospray: gravity_spheres_volume/dim_512/ao/real_timeospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/pathtracer/real_timey-cruncher: 1By-cruncher: 5By-cruncher: 500Monednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUospray-studio: 1 - 4K - 1 - Path Tracer - CPUospray-studio: 3 - 4K - 1 - Path Tracer - CPUospray-studio: 1 - 4K - 16 - Path Tracer - CPUospray-studio: 1 - 4K - 32 - Path Tracer - CPUospray-studio: 3 - 4K - 16 - Path Tracer - CPUospray-studio: 3 - 4K - 32 - Path Tracer - CPUospray-studio: 1 - 1080p - 1 - Path Tracer - CPUospray-studio: 3 - 1080p - 1 - Path Tracer - CPUospray-studio: 1 - 1080p - 16 - Path Tracer - CPUospray-studio: 1 - 1080p - 32 - Path Tracer - CPUospray-studio: 3 - 1080p - 16 - Path Tracer - CPUospray-studio: 3 - 1080p - 32 - Path Tracer - CPUtensorflow: CPU - 16 - ResNet-50tensorflow: CPU - 32 - ResNet-50tensorflow: CPU - 64 - ResNet-50openvino: Face Detection FP16 - CPUopenvino: Face Detection FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUSapphire RapidsEmerald Rapids AVX-512 On AVX-512 Off AVX-512 On AVX-512 Off3418.575136.7433466.394138.656124.0863152.5351131.8466270535.888235.243215.08324.91428.6042.293830.959889106214054330261679638483226268354570894231845766.1885.64103.28218.62137.09637.7347.01637.7847.003887.307.70412.86290.1514944.268.021528.7319.605759.7120.7326917.014.3820980.695.711894.4963.27742.5140.1947113.292.537556.5215.764381.7527.33144820.350.37151891.460.292090.44583.6182214.18288.567113.1300141.1111122.3348215523.658522.52119.225216.33836.6372.907896.5291035122616486383091960444460263309416683264917993447.2257.8468.4221.891355.33214.94139.44213.06140.641749.9817.1285.02352.086859.8617.46923.4732.453688.3232.512463.9848.4813852.888.651645.0372.89212.63140.688773.9813.662564.0911.691075.65111.6365805.381.75129576.780.575313.366212.5354975.695199.028151.4667200.9029174.3843324542.654541.610917.04095.23327.0582.7541331.7474988612125238251434532723189223301760533574714649.2672.0697.85236.14135.37747.4942.76747.7742.744765.956.69538.27237.3618384.836.951884.9916.958921.9714.3332234.043.9324769.105.152389.5553.511097.8229.0749132.992.4410161.3112.555105.9524.87114369.660.46123447.740.43392.568135.7033112.005124.480137.6227181.8347159.4832236523.410020.937410.566746.54033.6803.1092341.14893105514257336031715138787225279358572184242859639.0653.6066.0824.671287.11238.22134.18237.77134.442116.6415.1094.57337.698208.8815.581159.9327.564277.3129.912822.7145.3215101.298.461882.9767.93276.63115.359439.9213.552988.9110.691166.18109.7563672.971.7101244.000.63OpenBenchmarking.org

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1AVX-512 OnAVX-512 Off11002200330044005500SE +/- 12.90, N = 7SE +/- 17.21, N = 5SE +/- 3.87, N = 8SE +/- 0.74, N = 73418.582090.455313.373392.571. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1AVX-512 OnAVX-512 Off50100150200250SE +/- 0.52, N = 7SE +/- 0.69, N = 5SE +/- 0.15, N = 8SE +/- 0.03, N = 7136.7483.62212.54135.701. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2AVX-512 OnAVX-512 Off11002200330044005500SE +/- 17.85, N = 3SE +/- 5.42, N = 3SE +/- 49.58, N = 15SE +/- 24.22, N = 153466.392214.184975.703112.011. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2AVX-512 OnAVX-512 Off4080120160200SE +/- 0.71, N = 3SE +/- 0.22, N = 3SE +/- 1.98, N = 15SE +/- 0.97, N = 15138.6688.57199.03124.481. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: CrownAVX-512 OnAVX-512 Off306090120150SE +/- 0.36, N = 6SE +/- 0.27, N = 6SE +/- 0.55, N = 7SE +/- 0.23, N = 7124.09113.13151.47137.62

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian DragonAVX-512 OnAVX-512 Off4080120160200SE +/- 0.33, N = 7SE +/- 0.22, N = 7SE +/- 0.38, N = 8SE +/- 0.30, N = 7152.54141.11200.90181.83

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian Dragon ObjAVX-512 OnAVX-512 Off4080120160200SE +/- 0.26, N = 4SE +/- 0.21, N = 3SE +/- 0.52, N = 4SE +/- 0.19, N = 4131.85122.33174.38159.48

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 2.0.0Benchmark: vklBenchmarkCPU ISPCAVX-512 OnAVX-512 Off7001400210028003500SE +/- 3.84, N = 3SE +/- 4.04, N = 3SE +/- 30.78, N = 3SE +/- 6.36, N = 32705215532452365

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.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeAVX-512 OnAVX-512 Off1020304050SE +/- 0.20, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.07, N = 335.8923.6642.6523.41

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeAVX-512 OnAVX-512 Off918273645SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.16, N = 335.2422.5241.6120.94

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeAVX-512 OnAVX-512 Off48121620SE +/- 0.14483, N = 15SE +/- 0.13104, N = 15SE +/- 0.20744, N = 4SE +/- 0.07557, N = 1515.083209.2252117.0409010.56674

Y-Cruncher

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.2Pi Digits To Calculate: 1BAVX-512 OnAVX-512 Off246810SE +/- 0.003, N = 5SE +/- 0.007, N = 5SE +/- 0.021, N = 5SE +/- 0.026, N = 54.9146.3385.2336.540

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.2Pi Digits To Calculate: 5BAVX-512 OnAVX-512 Off816243240SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 328.6036.6427.0633.68

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.2Pi Digits To Calculate: 500MAVX-512 OnAVX-512 Off0.69951.3992.09852.7983.4975SE +/- 0.003, N = 7SE +/- 0.005, N = 7SE +/- 0.022, N = 7SE +/- 0.017, N = 72.2932.9072.7543.109

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 toolkit. Learn more via the OpenBenchmarking.org test page.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAVX-512 OnAVX-512 Off5001000150020002500SE +/- 10.80, N = 3SE +/- 6.64, N = 3SE +/- 31.97, N = 15SE +/- 54.32, N = 15830.96896.531331.742341.141. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

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.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off2004006008001000SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 2.03, N = 38891035749893

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off30060090012001500SE +/- 0.00, N = 3SE +/- 1.67, N = 3SE +/- 0.88, N = 3SE +/- 6.17, N = 3106212268861055

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off4K8K12K16K20KSE +/- 14.66, N = 3SE +/- 24.36, N = 3SE +/- 96.72, N = 3SE +/- 152.25, N = 514054164861212514257

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off8K16K24K32K40KSE +/- 114.21, N = 3SE +/- 150.37, N = 3SE +/- 74.75, N = 3SE +/- 274.25, N = 333026383092382533603

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off4K8K12K16K20KSE +/- 52.30, N = 3SE +/- 18.90, N = 3SE +/- 108.89, N = 15SE +/- 155.10, N = 716796196041434517151

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off10K20K30K40K50KSE +/- 213.70, N = 3SE +/- 123.86, N = 3SE +/- 167.00, N = 3SE +/- 82.99, N = 338483444603272338787

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off60120180240300SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3226263189225

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off70140210280350SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 2.41, N = 15268309223279

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off9001800270036004500SE +/- 1.86, N = 3SE +/- 5.17, N = 3SE +/- 7.00, N = 3SE +/- 23.43, N = 33545416630173585

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off2K4K6K8K10KSE +/- 17.64, N = 3SE +/- 8.69, N = 3SE +/- 27.24, N = 3SE +/- 60.01, N = 37089832660537218

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off11002200330044005500SE +/- 2.40, N = 3SE +/- 8.21, N = 3SE +/- 10.48, N = 3SE +/- 19.40, N = 34231491735744242

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUAVX-512 OnAVX-512 Off2K4K6K8K10KSE +/- 11.85, N = 3SE +/- 37.24, N = 3SE +/- 43.03, N = 3SE +/- 101.69, N = 38457993471468596

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 if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: ResNet-50AVX-512 OnAVX-512 Off1530456075SE +/- 0.31, N = 3SE +/- 0.28, N = 3SE +/- 0.41, N = 3SE +/- 0.20, N = 366.1847.2249.2639.06

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: ResNet-50AVX-512 OnAVX-512 Off20406080100SE +/- 0.22, N = 3SE +/- 0.36, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 385.6457.8472.0653.60

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: ResNet-50AVX-512 OnAVX-512 Off20406080100SE +/- 0.75, N = 3SE +/- 0.39, N = 3SE +/- 0.16, N = 3SE +/- 0.58, N = 3103.2868.4297.8566.08

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.

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off50100150200250SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3218.6221.89236.1424.671. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off30060090012001500SE +/- 0.07, N = 3SE +/- 4.01, N = 3SE +/- 0.03, N = 3SE +/- 2.43, N = 3137.091355.33135.371287.111. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off160320480640800SE +/- 0.66, N = 3SE +/- 0.08, N = 3SE +/- 0.82, N = 3SE +/- 0.31, N = 3637.73214.94747.49238.221. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off306090120150SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.18, N = 347.01139.4442.76134.181. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP32 - Device: CPUAVX-512 OnAVX-512 Off160320480640800SE +/- 0.38, N = 3SE +/- 0.26, N = 3SE +/- 1.03, N = 3SE +/- 0.09, N = 3637.78213.06747.77237.771. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Detection FP32 - Device: CPUAVX-512 OnAVX-512 Off306090120150SE +/- 0.03, N = 3SE +/- 0.18, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 347.00140.6442.74134.441. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off10002000300040005000SE +/- 1.67, N = 3SE +/- 0.53, N = 3SE +/- 1.59, N = 3SE +/- 1.56, N = 33887.301749.984765.952116.641. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off48121620SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 37.7017.126.6915.101. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off120240360480600SE +/- 2.15, N = 3SE +/- 0.06, N = 3SE +/- 0.53, N = 3SE +/- 0.10, N = 3412.8685.02538.2794.571. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off80160240320400SE +/- 1.52, N = 3SE +/- 0.22, N = 3SE +/- 0.24, N = 3SE +/- 0.38, N = 3290.15352.08237.36337.691. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16 - Device: CPUAVX-512 OnAVX-512 Off4K8K12K16K20KSE +/- 24.24, N = 3SE +/- 3.57, N = 3SE +/- 5.71, N = 3SE +/- 2.90, N = 314944.266859.8618384.838208.881. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16 - Device: CPUAVX-512 OnAVX-512 Off48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 38.0217.466.9515.581. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUAVX-512 OnAVX-512 Off400800120016002000SE +/- 1.66, N = 3SE +/- 0.24, N = 3SE +/- 1.06, N = 3SE +/- 0.98, N = 31528.73923.471884.991159.931. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUAVX-512 OnAVX-512 Off816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 319.6032.4516.9527.561. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off2K4K6K8K10KSE +/- 11.71, N = 3SE +/- 3.58, N = 3SE +/- 8.09, N = 3SE +/- 0.66, N = 35759.713688.328921.974277.311. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off816243240SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 320.7332.5114.3329.911. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off7K14K21K28K35KSE +/- 29.04, N = 3SE +/- 2.47, N = 3SE +/- 24.15, N = 3SE +/- 0.50, N = 326917.012463.9832234.042822.711. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off1122334455SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 34.3848.483.9345.321. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off5K10K15K20K25KSE +/- 25.65, N = 3SE +/- 9.76, N = 3SE +/- 20.85, N = 3SE +/- 8.51, N = 320980.6913852.8824769.1015101.291. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 35.718.655.158.461. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off5001000150020002500SE +/- 1.25, N = 3SE +/- 0.20, N = 3SE +/- 2.91, N = 3SE +/- 0.40, N = 31894.491645.032389.551882.971. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off1632486480SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 363.2772.8953.5167.931. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUAVX-512 OnAVX-512 Off2004006008001000SE +/- 1.33, N = 3SE +/- 1.82, N = 3SE +/- 1.48, N = 3SE +/- 0.41, N = 3742.51212.631097.82276.631. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUAVX-512 OnAVX-512 Off306090120150SE +/- 0.08, N = 3SE +/- 1.21, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 340.19140.6829.07115.351. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off11K22K33K44K55KSE +/- 54.78, N = 3SE +/- 3.89, N = 3SE +/- 80.54, N = 3SE +/- 3.35, N = 347113.298773.9849132.999439.921. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off48121620SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.5313.662.4413.551. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off2K4K6K8K10KSE +/- 6.80, N = 3SE +/- 1.81, N = 3SE +/- 3.93, N = 3SE +/- 0.72, N = 37556.522564.0910161.312988.911. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUAVX-512 OnAVX-512 Off48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 315.7611.6912.5510.691. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUAVX-512 OnAVX-512 Off11002200330044005500SE +/- 26.63, N = 3SE +/- 10.21, N = 15SE +/- 0.84, N = 3SE +/- 10.88, N = 74381.751075.655105.951166.181. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUAVX-512 OnAVX-512 Off20406080100SE +/- 0.15, N = 3SE +/- 1.05, N = 15SE +/- 0.01, N = 3SE +/- 1.00, N = 727.33111.6324.87109.751. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUAVX-512 OnAVX-512 Off30K60K90K120K150KSE +/- 735.86, N = 3SE +/- 51.71, N = 3SE +/- 898.88, N = 3SE +/- 52.05, N = 3144820.3565805.38114369.6663672.971. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUAVX-512 OnAVX-512 Off0.39380.78761.18141.57521.969SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 30.371.750.461.701. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off30K60K90K120K150KSE +/- 547.74, N = 3SE +/- 1069.55, N = 3SE +/- 940.07, N = 3SE +/- 496.38, N = 3151891.46129576.78123447.74101244.001. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUAVX-512 OnAVX-512 Off0.14180.28360.42540.56720.709SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.290.570.400.631. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

CPU Peak Freq (Highest CPU Core Frequency) Monitor

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgMegahertzCPU Peak Freq (Highest CPU Core Frequency) MonitorPhoronix Test Suite System MonitoringAVX-512 OnAVX-512 Off10002000300040005000Min: 500 / Avg: 2841.09 / Max: 3519Min: 800 / Avg: 2870.28 / Max: 5743Min: 800 / Avg: 2954.57 / Max: 3913Min: 500 / Avg: 3017.93 / Max: 3917

CPU Power Consumption Monitor

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringAVX-512 OnAVX-512 Off160320480640800Min: 72.02 / Avg: 574.93 / Max: 770.09Min: 67.03 / Avg: 598.76 / Max: 762.66Min: 99.6 / Avg: 589.71 / Max: 894.8Min: 103.63 / Avg: 583.08 / Max: 772.16

CPU Temperature Monitor

Sapphire RapidsEmerald RapidsOpenBenchmarking.orgCelsiusCPU Temperature MonitorPhoronix Test Suite System MonitoringAVX-512 OnAVX-512 Off1326395265Min: 22 / Avg: 48.5 / Max: 59Min: 25 / Avg: 51.04 / Max: 61Min: 30 / Avg: 54 / Max: 64Min: 29 / Avg: 53.19 / Max: 66

67 Results Shown

miniBUDE:
  OpenMP - BM1:
    GFInst/s
    Billion Interactions/s
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
Embree:
  Pathtracer ISPC - Crown
  Pathtracer ISPC - Asian Dragon
  Pathtracer ISPC - Asian Dragon Obj
OpenVKL
OSPRay:
  gravity_spheres_volume/dim_512/ao/real_time
  gravity_spheres_volume/dim_512/scivis/real_time
  gravity_spheres_volume/dim_512/pathtracer/real_time
Y-Cruncher:
  1B
  5B
  500M
oneDNN
OSPRay Studio:
  1 - 4K - 1 - Path Tracer - CPU
  3 - 4K - 1 - Path Tracer - CPU
  1 - 4K - 16 - Path Tracer - CPU
  1 - 4K - 32 - Path Tracer - CPU
  3 - 4K - 16 - Path Tracer - CPU
  3 - 4K - 32 - Path Tracer - CPU
  1 - 1080p - 1 - Path Tracer - CPU
  3 - 1080p - 1 - Path Tracer - CPU
  1 - 1080p - 16 - Path Tracer - CPU
  1 - 1080p - 32 - Path Tracer - CPU
  3 - 1080p - 16 - Path Tracer - CPU
  3 - 1080p - 32 - Path Tracer - CPU
TensorFlow:
  CPU - 16 - ResNet-50
  CPU - 32 - ResNet-50
  CPU - 64 - ResNet-50
OpenVINO:
  Face Detection FP16 - CPU:
    FPS
    ms
  Person Detection FP16 - CPU:
    FPS
    ms
  Person Detection FP32 - CPU:
    FPS
    ms
  Vehicle Detection FP16 - CPU:
    FPS
    ms
  Face Detection FP16-INT8 - CPU:
    FPS
    ms
  Face Detection Retail FP16 - CPU:
    FPS
    ms
  Road Segmentation ADAS FP16 - CPU:
    FPS
    ms
  Vehicle Detection FP16-INT8 - CPU:
    FPS
    ms
  Weld Porosity Detection FP16 - CPU:
    FPS
    ms
  Face Detection Retail FP16-INT8 - CPU:
    FPS
    ms
  Road Segmentation ADAS FP16-INT8 - CPU:
    FPS
    ms
  Machine Translation EN To DE FP16 - CPU:
    FPS
    ms
  Weld Porosity Detection FP16-INT8 - CPU:
    FPS
    ms
  Person Vehicle Bike Detection FP16 - CPU:
    FPS
    ms
  Handwritten English Recognition FP16 - CPU:
    FPS
    ms
  Age Gender Recognition Retail 0013 FP16 - CPU:
    FPS
    ms
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    FPS
    ms
CPU Peak Freq (Highest CPU Core Frequency) Monitor:
  Phoronix Test Suite System Monitoring:
    Megahertz
    Watts
    Celsius