AMD EPYC Genoa Memory Scaling

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2212240-NE-AMDEPYCGE62&grr.

WRF

Input: conus 2.5km

High Performance Conjugate Gradient

OpenVKL

Benchmark: vklBenchmark ISPC

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

NWChem

Input: C240 Buckyball

CockroachDB

Workload: KV, 10% Reads - Concurrency: 512

CockroachDB

Workload: KV, 95% Reads - Concurrency: 1024

CockroachDB

Workload: KV, 60% Reads - Concurrency: 512

CockroachDB

Workload: KV, 50% Reads - Concurrency: 1024

OSPRay

Benchmark: particle_volume/scivis/real_time

CockroachDB

Workload: KV, 50% Reads - Concurrency: 512

CockroachDB

Workload: KV, 95% Reads - Concurrency: 512

OSPRay

Benchmark: particle_volume/pathtracer/real_time

RELION

Test: Basic - Device: CPU

TensorFlow

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

LuxCoreRender

Scene: Danish Mood - Acceleration: CPU

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Standard

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenRadioss

Model: Bird Strike on Windshield

Apache Cassandra

Test: Writes

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OSPRay

Benchmark: particle_volume/ao/real_time

CockroachDB

Workload: KV, 60% Reads - Concurrency: 1024

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

Graph500

Scale: 26

Timed Linux Kernel Compilation

Build: allmodconfig

Timed Gem5 Compilation

Time To Compile

OSPRay

Benchmark: gravity_spheres_volume/dim_512/scivis/real_time

OSPRay

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

CockroachDB

Workload: KV, 10% Reads - Concurrency: 1024

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

Timed Node.js Compilation

Time To Compile

OpenRadioss

Model: INIVOL and Fluid Structure Interaction Drop Container

OpenVINO

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

OpenVINO

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

CockroachDB

Workload: MoVR - Concurrency: 512

CockroachDB

Workload: MoVR - Concurrency: 1024

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

nginx

Connections: 500

Timed Linux Kernel Compilation

Build: defconfig

simdjson

Throughput Test: TopTweet

OpenRadioss

Model: Bumper Beam

Blender

Blend File: Barbershop - Compute: CPU-Only

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

OpenVINO

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

OpenVINO

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

Timed LLVM Compilation

Build System: Ninja

simdjson

Throughput Test: DistinctUserID

simdjson

Throughput Test: PartialTweets

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenVINO

Model: Person Detection FP32 - Device: CPU

OpenVINO

Model: Person Detection FP32 - Device: CPU

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16 - Device: CPU

OpenVINO

Model: Face Detection FP16 - Device: CPU

libavif avifenc

Encoder Speed: 0

NAMD

ATPase Simulation - 327,506 Atoms

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

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

OpenVINO

Model: Weld Porosity Detection 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

Stargate Digital Audio Workstation

Sample Rate: 192000 - Buffer Size: 1024

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

simdjson

Throughput Test: Kostya

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

DaCapo Benchmark

Java Test: H2

simdjson

Throughput Test: LargeRandom

Build2

Time To Compile

Timed PHP Compilation

Time To Compile

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

Timed GDB GNU Debugger Compilation

Time To Compile

Neural Magic DeepSparse

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

Neural Magic DeepSparse

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

7-Zip Compression

Test: Decompression Rating

7-Zip Compression

Test: Compression Rating

nekRS

Input: TurboPipe Periodic

Stargate Digital Audio Workstation

Sample Rate: 96000 - Buffer Size: 1024

Timed Godot Game Engine Compilation

Time To Compile

libavif avifenc

Encoder Speed: 2

GPAW

Input: Carbon Nanotube

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

Rodinia

Test: OpenMP Streamcluster

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

oneDNN

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

NAS Parallel Benchmarks

Test / Class: IS.D

Blender

Blend File: Classroom - Compute: CPU-Only

Timed Apache Compilation

Time To Compile

Timed Mesa Compilation

Time To Compile

Liquid-DSP

Threads: 384 - Buffer Length: 256 - Filter Length: 57

Liquid-DSP

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

Intel Open Image Denoise

Run: RTLightmap.hdr.4096x4096

miniBUDE

Implementation: OpenMP - Input Deck: BM2

miniBUDE

Implementation: OpenMP - Input Deck: BM2

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

Kvazaar

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

Xmrig

Variant: Monero - Hash Count: 1M

NAS Parallel Benchmarks

Test / Class: CG.C

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

Xmrig

Variant: Wownero - Hash Count: 1M

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160

Blender

Blend File: BMW27 - Compute: CPU-Only

NAS Parallel Benchmarks

Test / Class: MG.C

NAS Parallel Benchmarks

Test / Class: SP.C

Kvazaar

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

ASTC Encoder

Preset: Exhaustive

Timed MPlayer Compilation

Time To Compile