Intel Xeon Platinum 8490H 1P vs. 2P Benchmarks

Sapphire Rapids launch day benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2301101-PTS-SPREXTRA34&grs.

BRL-CAD

VGR Performance Metric

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

OpenVINO

Model: Face Detection FP16 - Device: CPU

OpenVINO

Model: Face Detection FP16 - Device: CPU

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

NAS Parallel Benchmarks

Test / Class: SP.C

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

ACES DGEMM

Sustained Floating-Point Rate

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

Pennant

Test: leblancbig

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

Sysbench

Test: CPU

NAS Parallel Benchmarks

Test / Class: MG.C

Coremark

CoreMark Size 666 - Iterations Per Second

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

ASTC Encoder

Preset: Exhaustive

NAS Parallel Benchmarks

Test / Class: EP.D

ASKAP

Test: tConvolve MPI - Gridding

ASKAP

Test: tConvolve MPI - Degridding

Neural Magic DeepSparse

Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream

NAS Parallel Benchmarks

Test / Class: LU.C

OpenSSL

Algorithm: SHA256

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream

OpenSSL

Algorithm: RSA4096

Tachyon

Total Time

m-queens

Time To Solve

OpenSSL

Algorithm: RSA4096

C-Ray

Total Time - 4K, 16 Rays Per Pixel

High Performance Conjugate Gradient

RELION

Test: Basic - Device: CPU

7-Zip Compression

Test: Decompression Rating

Algebraic Multi-Grid Benchmark

Neural Magic DeepSparse

Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer

Blender

Blend File: Classroom - Compute: CPU-Only

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer

ASTC Encoder

Preset: Thorough

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer

NAS Parallel Benchmarks

Test / Class: BT.C

Neural Magic DeepSparse

Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream

OSPRay

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer

Liquid-DSP

Threads: 256 - Buffer Length: 256 - Filter Length: 57

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

Blender

Blend File: Barbershop - Compute: CPU-Only

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU

Blender

Blend File: BMW27 - Compute: CPU-Only

Liquid-DSP

Threads: 128 - Buffer Length: 256 - Filter Length: 57

WRF

Input: conus 2.5km

Blender

Blend File: Fishy Cat - Compute: CPU-Only

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

NAS Parallel Benchmarks

Test / Class: EP.C

LULESH

asmFish

1024 Hash Memory, 26 Depth

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

NAS Parallel Benchmarks

Test / Class: CG.C

Xmrig

Variant: Wownero - Hash Count: 1M

ASTC Encoder

Preset: Medium

Timed Linux Kernel Compilation

Build: allmodconfig

Xmrig

Variant: Monero - Hash Count: 1M

GraphicsMagick

Operation: Sharpen

NAS Parallel Benchmarks

Test / Class: SP.B

Intel Open Image Denoise

Run: RT.ldr_alb_nrm.3840x2160

Rodinia

Test: OpenMP LavaMD

Intel Open Image Denoise

Run: RTLightmap.hdr.4096x4096

OpenVINO

Model: Machine Translation EN To DE FP16 - Device: CPU

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160

OpenVINO

Model: Machine Translation EN To DE FP16 - Device: CPU

NAS Parallel Benchmarks

Test / Class: FT.C

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

Model: Person Detection FP32 - Device: CPU

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

Model: Person Detection FP32 - Device: CPU

GraphicsMagick

Operation: Enhanced

Embree

Binary: Pathtracer - Model: Asian Dragon

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

Embree

Binary: Pathtracer - Model: Crown

Timed LLVM Compilation

Build System: Ninja

POV-Ray

Trace Time

7-Zip Compression

Test: Compression Rating

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

Pennant

Test: sedovbig

Embree

Binary: Pathtracer ISPC - Model: Crown

Timed Node.js Compilation

Time To Compile

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

OpenRadioss

Model: INIVOL and Fluid Structure Interaction Drop Container

Timed Linux Kernel Compilation

Build: defconfig

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

NAS Parallel Benchmarks

Test / Class: IS.D

Neural Magic DeepSparse

Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream

Neural Magic DeepSparse

Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 57

Timed MPlayer Compilation

Time To Compile

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenRadioss

Model: Cell Phone Drop Test

Timed LLVM Compilation

Build System: Unix Makefiles

Graph500

Scale: 26

Timed Godot Game Engine Compilation

Time To Compile

LeelaChessZero

Backend: Eigen

Graph500

Scale: 26

NAMD

ATPase Simulation - 327,506 Atoms

Node.js Express HTTP Load Test

Build2

Time To Compile

OpenRadioss

Model: Bird Strike on Windshield

Timed Mesa Compilation

Time To Compile

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

Neural Magic DeepSparse

Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

Timed Gem5 Compilation

Time To Compile

Etcpak

Benchmark: Multi-Threaded - Configuration: ETC2

Zstd Compression

Compression Level: 19, Long Mode - Decompression Speed

Neural Magic DeepSparse

Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream

Zstd Compression

Compression Level: 19 - Decompression Speed

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

DaCapo Benchmark

Java Test: Jython

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Mesh Time

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

Zstd Compression

Compression Level: 19 - Compression Speed

NWChem

Input: C240 Buckyball

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream