AMD EPYC 5th Gen Benchmarks - Performance vs. Power Determinism

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2411056-NE-TURINPOWE18&sro&grs.

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

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

OpenVINO

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

OpenVINO

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

NAMD

Input: ATPase with 327,506 Atoms

OSPRay Studio

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

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

OpenVINO

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

OpenVINO

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

OpenVINO

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

NAS Parallel Benchmarks

Test / Class: SP.C

OpenSSL

Algorithm: RSA4096

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

Primesieve

Length: 1e13

Helsing

Digit Range: 14 digit

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OpenVINO

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

OpenSSL

Algorithm: RSA4096

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

OpenSSL

Algorithm: SHA512

RocksDB

Test: Random Read

OpenSSL

Algorithm: AES-256-GCM

QuantLib

Configuration: Multi-Threaded

OpenSSL

Algorithm: AES-128-GCM

OpenSSL

Algorithm: SHA256

OSPRay Studio

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

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

Speedb

Test: Random Read

OSPRay Studio

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

Blender

Blend File: Classroom - Compute: CPU-Only

OSPRay Studio

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

Blender

Blend File: BMW27 - Compute: CPU-Only

OSPRay Studio

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

Stockfish

Chess Benchmark

OpenVINO

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

ACES DGEMM

Sustained Floating-Point Rate

John The Ripper

Test: Blowfish

John The Ripper

Test: bcrypt

OpenSSL

Algorithm: ChaCha20

OpenSSL

Algorithm: ChaCha20-Poly1305

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

ASKAP

Test: tConvolve MPI - Gridding

ASTC Encoder

Preset: Exhaustive

OpenVINO

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

LULESH

Blender

Blend File: Barbershop - Compute: CPU-Only

Blender

Blend File: Fishy Cat - Compute: CPU-Only

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

ASKAP

Test: tConvolve MPI - Degridding

ASTC Encoder

Preset: Very Thorough

NAMD

Input: STMV with 1,066,628 Atoms

ONNX Runtime

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

Algebraic Multi-Grid Benchmark

NAS Parallel Benchmarks

Test / Class: LU.C

John The Ripper

Test: WPA PSK

Coremark

CoreMark Size 666 - Iterations Per Second

Pennant

Test: sedovbig

WRF

Input: conus 2.5km

NAS Parallel Benchmarks

Test / Class: IS.D

OpenRadioss

Model: Chrysler Neon 1M

miniBUDE

Implementation: OpenMP - Input Deck: BM1

miniBUDE

Implementation: OpenMP - Input Deck: BM1

OpenVINO

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

PyTorch

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

OpenVINO

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

Primesieve

Length: 1e12

OSPRay

Benchmark: particle_volume/scivis/real_time

OSPRay

Benchmark: particle_volume/ao/real_time

PyTorch

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

Embree

Binary: Pathtracer ISPC - Model: Crown

Blender

Blend File: Junkshop - Compute: CPU-Only

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

Timed Linux Kernel Compilation

Build: allmodconfig

ASTC Encoder

Preset: Thorough

OpenVKL

Benchmark: vklBenchmarkCPU ISPC

NAS Parallel Benchmarks

Test / Class: CG.C

7-Zip Compression

Test: Decompression Rating

PyTorch

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

NAS Parallel Benchmarks

Test / Class: MG.C

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

uvg266

Video Input: Bosphorus 4K - Video Preset: Slow

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Slow

Liquid-DSP

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

miniBUDE

Implementation: OpenMP - Input Deck: BM2

miniBUDE

Implementation: OpenMP - Input Deck: BM2

m-queens

Time To Solve

srsRAN Project

Test: PDSCH Processor Benchmark, Throughput Total

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

uvg266

Video Input: Bosphorus 4K - Video Preset: Medium

GraphicsMagick

Operation: Sharpen

GraphicsMagick

Operation: Enhanced

LuxCoreRender

Scene: DLSC - Acceleration: CPU

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon Obj

oneDNN

Harness: Deconvolution Batch shapes_3d - Engine: CPU

Liquid-DSP

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

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

Liquid-DSP

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