AMD 5th Gen EPYC 9005 Turin vs. NVIDIA GH200 Performance Benchmarks

AMD 5th Gen EPYC "Turin" Linux benchmarks by Michael Larabel for a future article on launch day of EPYC 9005 series Linux performance.

HTML result view exported from: https://openbenchmarking.org/result/2410298-NE-NVIDIAGRA46&sor&grr.

OpenVKL

Benchmark: vklBenchmarkCPU ISPC

RELION

Test: Basic - Device: CPU

NWChem

Input: C240 Buckyball

High Performance Conjugate Gradient

X Y Z: 144 144 144 - RT: 60

OpenSSL

Algorithm: RSA4096

OpenSSL

Algorithm: RSA4096

Speedb

Test: Read While Writing

BRL-CAD

VGR Performance Metric

SecureMark

Benchmark: SecureMark-TLS

Xmrig

Variant: GhostRider - Hash Count: 1M

Stockfish

Chess Benchmark

Timed Linux Kernel Compilation

Build: allmodconfig

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

OpenSSL

Algorithm: SHA512

LuxCoreRender

Scene: DLSC - Acceleration: CPU

OpenSSL

Algorithm: AES-128-GCM

OpenSSL

Algorithm: SHA256

OpenSSL

Algorithm: AES-256-GCM

OpenSSL

Algorithm: ChaCha20

OpenSSL

Algorithm: ChaCha20-Poly1305

OSPRay

Benchmark: particle_volume/scivis/real_time

TensorFlow

Device: CPU - Batch Size: 512 - Model: ResNet-50

Timed Node.js Compilation

Time To Compile

ClickHouse

100M Rows Hits Dataset, Third Run

ClickHouse

100M Rows Hits Dataset, Second Run

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

Timed Gem5 Compilation

Time To Compile

OSPRay

Benchmark: particle_volume/ao/real_time

Timed LLVM Compilation

Build System: Ninja

FFmpeg

Encoder: libx265 - Scenario: Platform

OpenRadioss

Model: Chrysler Neon 1M

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

ASKAP

Test: tConvolve MPI - Gridding

ASKAP

Test: tConvolve MPI - Degridding

Numpy Benchmark

Blender

Blend File: Barbershop - Compute: CPU-Only

Timed Godot Game Engine Compilation

Time To Compile

OpenRadioss

Model: INIVOL and Fluid Structure Interaction Drop Container

TensorFlow

Device: CPU - Batch Size: 256 - Model: ResNet-50

QMCPACK

Input: Li2_STO_ae

Zstd Compression

Compression Level: 19, Long Mode - Decompression Speed

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

OSPRay Studio

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

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

Zstd Compression

Compression Level: 19 - Decompression Speed

Zstd Compression

Compression Level: 19 - Compression Speed

libavif avifenc

Encoder Speed: 0

Memcached

Set To Get Ratio: 1:100

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

Model: Person Detection FP16 - Device: CPU

OSPRay Studio

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

OpenVINO

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

OpenVINO

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

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

Model: Road Segmentation ADAS FP16-INT8 - Device: CPU

OpenVINO

Model: Road Segmentation ADAS FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection Retail FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection Retail FP16-INT8 - Device: CPU

OpenVINO

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

OpenVINO

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

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OSPRay Studio

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

OpenVINO

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

OpenVINO

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

GraphicsMagick

Operation: Sharpen

Speedb

Test: Random Read

GraphicsMagick

Operation: Enhanced

GraphicsMagick

Operation: Noise-Gaussian

Liquid-DSP

Threads: 1 - Buffer Length: 256 - Filter Length: 512

RocksDB

Test: Random Read

NAMD

Input: STMV with 1,066,628 Atoms

OSPRay Studio

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

OSPRay Studio

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

Helsing

Digit Range: 14 digit

OSPRay Studio

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

QuantLib

Configuration: Multi-Threaded

miniBUDE

Implementation: OpenMP - Input Deck: BM2

miniBUDE

Implementation: OpenMP - Input Deck: BM2

NAMD

Input: ATPase with 327,506 Atoms

WebP Image Encode

Encode Settings: Quality 100, Lossless, Highest Compression

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

NAS Parallel Benchmarks

Test / Class: SP.C

Liquid-DSP

Threads: 256 - Buffer Length: 256 - Filter Length: 512

John The Ripper

Test: WPA PSK

Liquid-DSP

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

Liquid-DSP

Threads: 128 - Buffer Length: 256 - Filter Length: 512

Liquid-DSP

Threads: 256 - Buffer Length: 256 - Filter Length: 32

John The Ripper

Test: Blowfish

John The Ripper

Test: bcrypt

Liquid-DSP

Threads: 1 - Buffer Length: 256 - Filter Length: 32

Blender

Blend File: Classroom - Compute: CPU-Only

7-Zip Compression

Test: Decompression Rating

7-Zip Compression

Test: Compression Rating

ASTC Encoder

Preset: Exhaustive

OpenRadioss

Model: Cell Phone Drop Test

ASTC Encoder

Preset: Very Thorough

Algebraic Multi-Grid Benchmark

Primesieve

Length: 1e13

NAS Parallel Benchmarks

Test / Class: LU.C

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Total

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

CloverLeaf

Input: clover_bm64_short

NAS Parallel Benchmarks

Test / Class: IS.D

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Thread

GPAW

Input: Carbon Nanotube

Pennant

Test: sedovbig

Coremark

CoreMark Size 666 - Iterations Per Second

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Mesh Time

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 4K

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

uvg266

Video Input: Bosphorus 4K - Video Preset: Slow

uvg266

Video Input: Bosphorus 4K - Video Preset: Medium

NAS Parallel Benchmarks

Test / Class: CG.C

Blender

Blend File: Junkshop - Compute: CPU-Only

PyBench

Total For Average Test Times

Blender

Blend File: Fishy Cat - Compute: CPU-Only

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Slow

Blender

Blend File: BMW27 - Compute: CPU-Only

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

SVT-AV1

Encoder Mode: Preset 13 - Input: Bosphorus 4K

Intel Open Image Denoise

Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only

ASTC Encoder

Preset: Thorough

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon Obj

Pennant

Test: leblancbig

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

uvg266

Video Input: Bosphorus 4K - Video Preset: Very Fast

miniFE

Problem Size: Small

TensorFlow

Device: CPU - Batch Size: 256 - Model: AlexNet

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only

NAS Parallel Benchmarks

Test / Class: MG.C

Intel Open Image Denoise

Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only