EPYC 7F72 Kernel Testing

AMD EPYC 7F72 24-Core testing with a ASRockRack EPYCD8 (P2.10 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2009146-FI-EPYC7F72K42&grr.

PlaidML

FP16: No - Mode: Inference - Network: NASNer Large - Device: CPU

Numenta Anomaly Benchmark

Detector: EXPoSE

Renaissance

Test: Akka Unbalanced Cobwebbed Tree

Renaissance

Test: Savina Reactors.IO

Java Gradle Build

Gradle Build: Reactor

Renaissance

Test: Apache Spark PageRank

PlaidML

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU

Renaissance

Test: In-Memory Database Shootout

Apache Cassandra

Test: Reads

BRL-CAD

VGR Performance Metric

Renaissance

Test: Apache Spark ALS

MariaDB

Clients: 512

MariaDB

Clients: 256

Memcached mcperf

Method: Add - Connections: 16

Memcached mcperf

Method: Set - Connections: 16

LeelaChessZero

Backend: Eigen

Memcached mcperf

Method: Replace - Connections: 16

Memcached mcperf

Method: Append - Connections: 16

Memcached mcperf

Method: Prepend - Connections: 16

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU

OpenVKL

Benchmark: vklBenchmark

LeelaChessZero

Backend: BLAS

AI Benchmark Alpha

Device AI Score

AI Benchmark Alpha

Device Training Score

AI Benchmark Alpha

Device Inference Score

Renaissance

Test: Random Forest

MariaDB

Clients: 128

ECP-CANDLE

Benchmark: P3B2

Blender

Blend File: Barbershop - Compute: CPU-Only

Timed LLVM Compilation

Time To Compile

Memcached mcperf

Method: Delete - Connections: 16

Memcached mcperf

Method: Get - Connections: 16

PlaidML

FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

ECP-CANDLE

Benchmark: P3B1

Renaissance

Test: Genetic Algorithm Using Jenetics + Futures

Blender

Blend File: Classroom - Compute: CPU-Only

G'MIC

Test: 2D Function Plotting, 1000 Times

CP2K Molecular Dynamics

Fayalite-FIST Data

MariaDB

Clients: 64

PlaidML

FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU

Apache Siege

Concurrent Users: 250

High Performance Conjugate Gradient

Renaissance

Test: Scala Dotty

Schbench

Message Threads: 2 - Workers Per Message Thread: 6

TensorFlow Lite

Model: NASNet Mobile

Schbench

Message Threads: 2 - Workers Per Message Thread: 2

Schbench

Message Threads: 2 - Workers Per Message Thread: 4

Apache Siege

Concurrent Users: 200

Memcached mcperf

Method: Prepend - Connections: 1

Memcached mcperf

Method: Delete - Connections: 1

Renaissance

Test: Apache Spark Bayes

Ethr

Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 16

Stress-NG

Test: CPU Cache

Stress-NG

Test: Forking

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

PostgreSQL pgbench

Scaling: Buffer Test - Test: Single Thread - Mode: Read Only

PostgreSQL pgbench

Scaling: Buffer Test - Test: Single Thread - Mode: Read Write

Memcached mcperf

Method: Add - Connections: 1

Mlpack Benchmark

Benchmark: scikit_qda

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: CPU

Rodinia

Test: OpenMP LavaMD

Rodinia

Test: OpenMP HotSpot3D

G'MIC

Test: 3D Elevated Function In Random Colors, 100 Times

Geekbench

Test: CPU Multi Core - Horizon Detection

Geekbench

Test: CPU Multi Core - Face Detection

Geekbench

Test: CPU Multi Core - Gaussian Blur

Geekbench

Test: CPU Multi Core

Schbench

Message Threads: 4 - Workers Per Message Thread: 2

Stress-NG

Test: Glibc C String Functions

YafaRay

Total Time For Sample Scene

Blender

Blend File: Fishy Cat - Compute: CPU-Only

Schbench

Message Threads: 4 - Workers Per Message Thread: 8

PyPerformance

Benchmark: raytrace

ASTC Encoder

Preset: Exhaustive

C-Blosc

Compressor: blosclz

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: CPU

perf-bench

Benchmark: Memcpy 1MB

Apache Cassandra

Test: Writes

Numenta Anomaly Benchmark

Detector: Earthgecko Skyline

Renaissance

Test: Twitter HTTP Requests

Memcached mcperf

Method: Replace - Connections: 1

Schbench

Message Threads: 4 - Workers Per Message Thread: 4

PyPerformance

Benchmark: python_startup

MariaDB

Clients: 32

SQLite Speedtest

Timed Time - Size 1,000

KeyDB

Schbench

Message Threads: 4 - Workers Per Message Thread: 6

Schbench

Message Threads: 2 - Workers Per Message Thread: 8

Memcached mcperf

Method: Append - Connections: 1

Build2

Time To Compile

pmbench

Concurrent Worker Threads: 2 - Read-Write Ratio: 50%

Blender

Blend File: BMW27 - Compute: CPU-Only

TensorFlow Lite

Model: Inception V4

Stress-NG

Test: Context Switching

Hackbench

Count: 16 - Type: Thread

Mlpack Benchmark

Benchmark: scikit_ica

PyPerformance

Benchmark: 2to3

Node.js Express HTTP Load Test

Hackbench

Count: 16 - Type: Process

Mlpack Benchmark

Benchmark: scikit_linearridgeregression

libavif avifenc

Encoder Speed: 0

NGINX Benchmark

Static Web Page Serving

TensorFlow Lite

Model: Inception ResNet V2

pmbench

Concurrent Worker Threads: 4 - Read-Write Ratio: 50%

pmbench

Concurrent Worker Threads: 1 - Read-Write Ratio: 100% Writes

pmbench

Concurrent Worker Threads: 1 - Read-Write Ratio: 50%

pmbench

Concurrent Worker Threads: 8 - Read-Write Ratio: 50%

pmbench

Concurrent Worker Threads: 2 - Read-Write Ratio: 100% Writes

pmbench

Concurrent Worker Threads: 8 - Read-Write Ratio: 100% Writes

pmbench

Concurrent Worker Threads: 4 - Read-Write Ratio: 100% Writes

pmbench

Concurrent Worker Threads: 2 - Read-Write Ratio: 100% Reads

pmbench

Concurrent Worker Threads: 8 - Read-Write Ratio: 100% Reads

pmbench

Concurrent Worker Threads: 4 - Read-Write Ratio: 100% Reads

pmbench

Concurrent Worker Threads: 1 - Read-Write Ratio: 100% Reads

TensorFlow Lite

Model: SqueezeNet

TensorFlow Lite

Model: Mobilenet Quant

TensorFlow Lite

Model: Mobilenet Float

Himeno Benchmark

Poisson Pressure Solver

Coremark

CoreMark Size 666 - Iterations Per Second

Geekbench

Test: CPU Single Core - Horizon Detection

Geekbench

Test: CPU Single Core - Face Detection

Geekbench

Test: CPU Single Core - Gaussian Blur

Geekbench

Test: CPU Single Core

Git

Time To Complete Common Git Commands

Hackbench

Count: 32 - Type: Process

Rodinia

Test: OpenMP Leukocyte