AMD EPYC 9005 Benchmarks - Performance vs. Power Determinism

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2411056-NE-TURINPOWE15&sro&grs.

miniBUDE

Implementation: OpenMP - Input Deck: BM1

miniBUDE

Implementation: OpenMP - Input Deck: BM1

OSPRay Studio

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

OpenVINO

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

OSPRay Studio

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

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

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

oneDNN

Harness: Deconvolution Batch shapes_3d - Engine: CPU

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

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

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

Primesieve

Length: 1e13

miniBUDE

Implementation: OpenMP - Input Deck: BM2

miniBUDE

Implementation: OpenMP - Input Deck: BM2

OpenVINO

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

OpenSSL

Algorithm: RSA4096

Helsing

Digit Range: 14 digit

NAS Parallel Benchmarks

Test / Class: CG.C

7-Zip Compression

Test: Decompression Rating

OSPRay Studio

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

OSPRay Studio

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

Primesieve

Length: 1e12

OSPRay Studio

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

OSPRay Studio

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

ASTC Encoder

Preset: Very Thorough

ASTC Encoder

Preset: Exhaustive

m-queens

Time To Solve

OpenVINO

Model: Person Re-Identification Retail FP16 - Device: CPU

ASTC Encoder

Preset: Thorough

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

RocksDB

Test: Random Read

OpenSSL

Algorithm: SHA512

OpenSSL

Algorithm: RSA4096

Blender

Blend File: BMW27 - Compute: CPU-Only

ACES DGEMM

Sustained Floating-Point Rate

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

OpenVINO

Model: Handwritten English Recognition FP16-INT8 - Device: CPU

John The Ripper

Test: Blowfish

John The Ripper

Test: bcrypt

OpenSSL

Algorithm: AES-256-GCM

OpenSSL

Algorithm: AES-128-GCM

GraphicsMagick

Operation: Enhanced

Blender

Blend File: Classroom - Compute: CPU-Only

QuantLib

Configuration: Multi-Threaded

OpenSSL

Algorithm: SHA256

Blender

Blend File: Fishy Cat - Compute: CPU-Only

Speedb

Test: Random Read

GraphicsMagick

Operation: Sharpen

Embree

Binary: Pathtracer ISPC - Model: Crown

Liquid-DSP

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

Coremark

CoreMark Size 666 - Iterations Per Second

Blender

Blend File: Barbershop - Compute: CPU-Only

OpenVINO

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

srsRAN Project

Test: PDSCH Processor Benchmark, Throughput Total

TensorFlow

Device: CPU - Batch Size: 512 - Model: AlexNet

Liquid-DSP

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

OpenVKL

Benchmark: vklBenchmarkCPU ISPC

Liquid-DSP

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

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon Obj

OSPRay

Benchmark: particle_volume/scivis/real_time

OSPRay

Benchmark: particle_volume/ao/real_time

John The Ripper

Test: WPA PSK

Blender

Blend File: Junkshop - Compute: CPU-Only

OpenVINO

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

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

uvg266

Video Input: Bosphorus 4K - Video Preset: Slow

NAS Parallel Benchmarks

Test / Class: SP.C

uvg266

Video Input: Bosphorus 4K - Video Preset: Medium

Y-Cruncher

Pi Digits To Calculate: 1B

ASKAP

Test: tConvolve MPI - Degridding

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

ASTC Encoder

Preset: Medium

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Slow

ASKAP

Test: tConvolve MPI - Gridding

Parallel BZIP2 Compression

FreeBSD-13.0-RELEASE-amd64-memstick.img Compression

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

NAS Parallel Benchmarks

Test / Class: IS.D

TensorFlow

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

NAS Parallel Benchmarks

Test / Class: LU.C

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

7-Zip Compression

Test: Compression Rating

Intel Open Image Denoise

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

OpenRadioss

Model: Chrysler Neon 1M

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

Intel Open Image Denoise

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

Intel Open Image Denoise

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

NAMD

Input: ATPase with 327,506 Atoms

NAS Parallel Benchmarks

Test / Class: MG.C

OpenVINO

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

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Mesh Time

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

GPAW

Input: Carbon Nanotube

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

LuxCoreRender

Scene: LuxCore Benchmark - Acceleration: CPU

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard

PyTorch

Device: CPU - Batch Size: 256 - Model: ResNet-152

PyTorch

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

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

Timed LLVM Compilation

Build System: Ninja

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

PyTorch

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

WRF

Input: conus 2.5km

NAMD

Input: STMV with 1,066,628 Atoms

GraphicsMagick

Operation: Noise-Gaussian

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OpenRadioss

Model: INIVOL and Fluid Structure Interaction Drop Container

Timed Node.js Compilation

Time To Compile

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

Memcached

Set To Get Ratio: 1:100

Kvazaar

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

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

Timed Linux Kernel Compilation

Build: allmodconfig

Timed Mesa Compilation

Time To Compile

OpenVINO

Model: Person Re-Identification Retail FP16 - Device: CPU

OpenVINO

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

ClickHouse

100M Rows Hits Dataset, Third Run

VVenC

Video Input: Bosphorus 4K - Video Preset: Faster

ClickHouse

100M Rows Hits Dataset, Second Run

TensorFlow

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

OpenVINO

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

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

x265

Video Input: Bosphorus 4K

TensorFlow

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

VVenC

Video Input: Bosphorus 4K - Video Preset: Fast

Timed ImageMagick Compilation

Time To Compile