AMD EPYC Zen 5 SMT Comparison

AMD EPYC 9575F 1P SMT comparison benchmarks by Michael Larabel for a future article. Fresh tests repeated with SMT on/off from SMCI BIOS toggle.

HTML result view exported from: https://openbenchmarking.org/result/2501314-NE-AMDEPYCZE72&grs&sro.

OpenVINO

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

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

GraphicsMagick

Operation: Resizing

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

RocksDB

Test: Update Random

OpenVINO

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

OpenVINO

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

Speedb

Test: Update Random

7-Zip Compression

Test: Decompression Rating

Embree

Binary: Pathtracer ISPC - Model: Crown

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon Obj

John The Ripper

Test: WPA PSK

RocksDB

Test: Read While Writing

John The Ripper

Test: bcrypt

John The Ripper

Test: Blowfish

OpenSSL

Algorithm: ChaCha20-Poly1305

Coremark

CoreMark Size 666 - Iterations Per Second

Stress-NG

Test: Hyperbolic Trigonometric Math

Stress-NG

Test: CPU Stress

OpenSSL

Algorithm: ChaCha20

LuxCoreRender

Scene: Rainbow Colors and Prism - Acceleration: CPU

LuxCoreRender

Scene: LuxCore Benchmark - Acceleration: CPU

LuxCoreRender

Scene: Danish Mood - Acceleration: CPU

Stress-NG

Test: Vector Math

OSPRay Studio

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

OSPRay Studio

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

OSPRay Studio

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

Memcached

Set To Get Ratio: 1:100

OSPRay Studio

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

GraphicsMagick

Operation: Enhanced

OSPRay Studio

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

OpenSSL

Algorithm: SHA256

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

OSPRay Studio

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

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Only

OpenVKL

Benchmark: vklBenchmarkCPU ISPC

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Only

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

John The Ripper

Test: MD5

7-Zip Compression

Test: Compression Rating

Blender

Blend File: Junkshop - Compute: CPU-Only

Blender

Blend File: Barbershop - Compute: CPU-Only

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Write

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency

LuxCoreRender

Scene: DLSC - Acceleration: CPU

Stress-NG

Test: Context Switching

Blender

Blend File: Classroom - Compute: CPU-Only

Blender

Blend File: BMW27 - Compute: CPU-Only

RocksDB

Test: Read Random Write Random

GraphicsMagick

Operation: Noise-Gaussian

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Write

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency

NAS Parallel Benchmarks

Test / Class: SP.B

Stress-NG

Test: Integer Math

Liquid-DSP

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

Stress-NG

Test: AVX-512 VNNI

OpenVINO

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

OpenSSL

Algorithm: RSA4096

nginx

Connections: 500

Timed Linux Kernel Compilation

Build: allmodconfig

OpenVINO

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

Stress-NG

Test: CPU Cache

NAMD

Input: ATPase with 327,506 Atoms

NAMD

Input: STMV with 1,066,628 Atoms

Timed LLVM Compilation

Build System: Ninja

John The Ripper

Test: HMAC-SHA512

GraphicsMagick

Operation: Swirl

OpenVINO

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

NAS Parallel Benchmarks

Test / Class: MG.C

OpenVINO

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

nginx

Connections: 1000

NAS Parallel Benchmarks

Test / Class: LU.C

Memcached

Set To Get Ratio: 1:10

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Total

Tachyon

Total Time

OpenVINO

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

Stress-NG

Test: Integer Bit Operations

Timed Node.js Compilation

Time To Compile

srsRAN Project

Test: PDSCH Processor Benchmark, Throughput Total

OpenSSL

Algorithm: SHA512

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Slow

VVenC

Video Input: Bosphorus 4K - Video Preset: Faster

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

GraphicsMagick

Operation: HWB Color Space

VVenC

Video Input: Bosphorus 4K - Video Preset: Fast

uvg266

Video Input: Bosphorus 4K - Video Preset: Slow

Timed Eigen Compilation

Time To Compile

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

Timed Gem5 Compilation

Time To Compile

NAS Parallel Benchmarks

Test / Class: BT.C

Primesieve

Length: 1e13

uvg266

Video Input: Bosphorus 4K - Video Preset: Medium

OpenSSL

Algorithm: AES-256-GCM

Kvazaar

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

uvg266

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

OpenVINO

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

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

ACES DGEMM

Sustained Floating-Point Rate

uvg266

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

Stress-NG

Test: NUMA

Kvazaar

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

SVT-AV1

Encoder Mode: Preset 5 - Input: Bosphorus 4K

NAS Parallel Benchmarks

Test / Class: SP.C

ClickHouse

100M Rows Hits Dataset, Second Run

OpenVINO

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

Kvazaar

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

Stress-NG

Test: Memory Copying

GraphicsMagick

Operation: Rotate

Llama.cpp

Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Text Generation 128

OpenVINO

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

Speedb

Test: Random Read

C-Ray

Resolution: 5K - Rays Per Pixel: 16

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

SVT-AV1

Encoder Mode: Preset 3 - Input: Bosphorus 4K

C-Ray

Resolution: 4K - Rays Per Pixel: 16

RocksDB

Test: Random Read

ClickHouse

100M Rows Hits Dataset, Third Run

GraphicsMagick

Operation: Sharpen

Laghos

Test: Triple Point Problem

NAS Parallel Benchmarks

Test / Class: IS.D

Speedb

Test: Read Random Write Random

Laghos

Test: Sedov Blast Wave, ube_922_hex.mesh

OpenVINO GenAI

Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU

OpenSSL

Algorithm: AES-128-GCM

Rustls

Benchmark: handshake-resume - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

Rustls

Benchmark: handshake-ticket - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

Rustls

Benchmark: handshake-ticket - Suite: TLS13_CHACHA20_POLY1305_SHA256

OpenVINO

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

OpenSSL

Algorithm: RSA4096

uvg266

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

Palabos

Grid Size: 500

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

Rustls

Benchmark: handshake-resume - Suite: TLS13_CHACHA20_POLY1305_SHA256

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 512

Rustls

Benchmark: handshake-ticket - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

FinanceBench

Benchmark: Repo OpenMP

Rustls

Benchmark: handshake-resume - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

Llama.cpp

Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Text Generation 128

Llama.cpp

Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Text Generation 128

FinanceBench

Benchmark: Bonds OpenMP

Rustls

Benchmark: handshake - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

Rustls

Benchmark: handshake - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

Rustls

Benchmark: handshake - Suite: TLS13_CHACHA20_POLY1305_SHA256

C-Ray

CPU Temperature Monitor

C-Ray

CPU Power Consumption Monitor

C-Ray

CPU Peak Freq (Highest CPU Core Frequency) Monitor

C-Ray

CPU Temperature Monitor

C-Ray

CPU Power Consumption Monitor