AMD EPYC vs. Intel Xeon vs. POWER9 Linux Benchmarks
AMD EPYC vs. Intel Xeon open-source server benchmarks on Ubuntu Linux. Latest as of November 2018. Tests by Michael Larabel for a future article.
HTML result view exported from: https://openbenchmarking.org/result/1811168-SK-EPYCPOWER73&grs.
Blender
Blend File: Classroom - Compute: CPU-Only
John The Ripper
Test: Traditional DES
Sysbench
Test: CPU
John The Ripper
Test: Blowfish
Rodinia
Test: OpenMP LavaMD
Blender
Blend File: Classroom - Compute: CPU-Only
Stress-NG
Test: Socket Activity
POV-Ray
Trace Time
NAMD
ATPase Simulation - 327,506 Atoms
Stress-NG
Test: Matrix Math
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
ebizzy
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
Smallpt
Global Illumination Renderer; 128 Samples
Stockfish
Total Time
asmFish
1024 Hash Memory, 26 Depth
Blender
Blend File: Fishy Cat - Compute: CPU-Only
VP9 libvpx Encoding
vpxenc
Blender
Blend File: Fishy Cat - Compute: CPU-Only
Stress-NG
Test: CPU Stress
Primesieve
1e12 Prime Number Generation
Aircrack-ng
OpenSSL
RSA 4096-bit Performance
NAS Parallel Benchmarks
Test / Class: EP.C
Blender
Blend File: BMW27 - Compute: CPU-Only
Blender
Blend File: BMW27 - Compute: CPU-Only
NAS Parallel Benchmarks
Test / Class: LU.A
Tachyon
Total Time
Blender
Blend File: Barbershop - Compute: CPU-Only
NAS Parallel Benchmarks
Test / Class: LU.C
Blender
Blend File: Barbershop - Compute: CPU-Only
7-Zip Compression
Compress Speed Test
Chaos Group V-RAY
Mode: CPU
IndigoBench
Scene: Bedroom
Timed Linux Kernel Compilation
Time To Compile
SciMark
Computational Test: Monte Carlo
Timed LLVM Compilation
Time To Compile
x265
H.265 Video Encoding
IndigoBench
Scene: Supercar
FLAC Audio Encoding
WAV To FLAC
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
Y-Cruncher
Calculating 500M Pi Digits
Parboil
Test: OpenMP CUTCP
Mencoder
AVI To LAVC
x264
H.264 Video Encoding
Stream
Type: Triad
Stream
Type: Add
Stream
Type: Scale
SciMark
Computational Test: Fast Fourier Transform
Bork File Encrypter
File Encryption Time
SciMark
Computational Test: Sparse Matrix Multiply
MBW
Test: Memory Copy - Array Size: 1024 MiB
MBW
Test: Memory Copy - Array Size: 512 MiB
lzbench
Test: Libdeflate 1 - Process: Decompression
MBW
Test: Memory Copy - Array Size: 8192 MiB
MBW
Test: Memory Copy - Array Size: 4096 MiB
MBW
Test: Memory Copy - Array Size: 128 MiB
Memcached mcperf
Method: Append
SciMark
Computational Test: Dense LU Matrix Factorization
Timed ImageMagick Compilation
Time To Compile
FFTW
Build: Float + SSE - Size: 1D FFT Size 1024
Stream
Type: Copy
LuaJIT
Test: Fast Fourier Transform
Zstd Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19
CacheBench
Test: Read
FFTW
Build: Float + SSE - Size: 1D FFT Size 2048
Optcarrot
Optimized Benchmark
FFTW
Build: Float + SSE - Size: 1D FFT Size 4096
R Benchmark
FFTW
Build: Float + SSE - Size: 1D FFT Size 512
glibc bench
Benchmark: sqrt
Java SciMark
Computational Test: Dense LU Matrix Factorization
CacheBench
Test: Write
GraphicsMagick
Operation: Noise-Gaussian
Timed HMMer Search
Pfam Database Search
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 16 - Solver Time
Timed GCC Compilation
Time To Compile
Tinymembench
Standard Memset
Tinymembench
Standard Memcpy
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 16
Swet
Average
GraphicsMagick
Operation: Sharpen
Java SciMark
Computational Test: Sparse Matrix Multiply
OSBench
Test: Create Files
Rust Prime Benchmark
Prime Number Test To 200,000,000
m-queens
Time To Solve
Stress-NG
Test: Crypto
Stress-NG
Test: Vector Math
N-Queens
Elapsed Time
MariaDB
Clients: 64
C-Ray
Total Time - 4K, 16 Rays Per Pixel
CacheBench
Test: Read / Modify / Write
SciMark
Computational Test: Composite
Redis
Test: LPUSH
Compile Bench
Test: Compile
Bullet Physics Engine
Test: 1000 Convex
GIMP
Test: auto-levels
Darktable
Test: Masskrug - Acceleration: CPU-only
Java SciMark
Computational Test: Composite
LuaJIT
Test: Dense LU Matrix Factorization
FFTW
Build: Float + SSE - Size: 1D FFT Size 256
Bullet Physics Engine
Test: Raytests
Sudokut
Total Time
GraphicsMagick
Operation: Enhanced
glibc bench
Benchmark: ffsll
Numpy Benchmark
PHPBench
PHP Benchmark Suite
glibc bench
Benchmark: ffs
Java SciMark
Computational Test: Jacobi Successive Over-Relaxation
RAR Compression
Linux Source Tree Archiving To RAR
FFTW
Build: Stock - Size: 1D FFT Size 32
FFmpeg
H.264 HD To NTSC DV
Rust Mandelbrot
Time To Complete Serial/Parallel Mandelbrot
lzbench
Test: Libdeflate 1 - Process: Compression
FFTW
Build: Stock - Size: 1D FFT Size 64
FFTW
Build: Float + SSE - Size: 1D FFT Size 32
Bullet Physics Engine
Test: Convex Trimesh
Cryptsetup
PBKDF2-sha512
Timed PHP Compilation
Time To Compile
DaCapo Benchmark
Java Test: Jython
Node.js Octane Benchmark
AOBench
Size: 2048 x 2048 - Total Time
Java SciMark
Computational Test: Monte Carlo
lzbench
Test: Brotli 0 - Process: Decompression
GIMP
Test: rotate
Bullet Physics Engine
Test: 1000 Stack
eSpeak Speech Engine
Text-To-Speech Synthesis
Redis
Test: SET
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 32 - Solver Time
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 32
PyBench
Total For Average Test Times
LAME MP3 Encoding
WAV To MP3
FFTW
Build: Float + SSE - Size: 2D FFT Size 1024
Compile Bench
Test: Read Compiled Tree
Redis
Test: GET
GNU MPC
Multi-Precision Benchmark
FFTW
Build: Float + SSE - Size: 2D FFT Size 32
Java SciMark
Computational Test: Fast Fourier Transform
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB
lzbench
Test: XZ 0 - Process: Compression
Bullet Physics Engine
Test: Prim Trimesh
glibc bench
Benchmark: sin
Bullet Physics Engine
Test: 136 Ragdolls
glibc bench
Benchmark: cos
PostgreSQL pgbench
Scaling: Buffer Test - Test: Single Thread - Mode: Read Only
LuaJIT
Test: Composite
Hierarchical INTegration
Test: FLOAT
libjpeg-turbo tjbench
Test: Decompression Throughput
TSCP
AI Chess Performance
dcraw
RAW To PPM Image Conversion
glibc bench
Benchmark: pthread_once
GNU Octave Benchmark
FFTW
Build: Float + SSE - Size: 2D FFT Size 2048
OSBench
Test: Memory Allocations
Redis
Test: SADD
BYTE Unix Benchmark
Computational Test: Dhrystone 2
SciMark
Computational Test: Jacobi Successive Over-Relaxation
Bullet Physics Engine
Test: 3000 Fall
glibc bench
Benchmark: tanh
FFTW
Build: Float + SSE - Size: 1D FFT Size 128
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 16
GraphicsMagick
Operation: Swirl
Minion
Benchmark: Quasigroup
Apache Benchmark
Static Web Page Serving
GIMP
Test: resize
FFTW
Build: Stock - Size: 2D FFT Size 32
FFTW
Build: Stock - Size: 1D FFT Size 128
Cryptsetup
PBKDF2-whirlpool
GNU GMP GMPbench
Total Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 16 - Solver Time
Crafty
Elapsed Time
FFTW
Build: Stock - Size: 2D FFT Size 64
LuaJIT
Test: Sparse Matrix Multiply
Timed Apache Compilation
Time To Compile
FFTW
Build: Float + SSE - Size: 2D FFT Size 4096
Tesseract OCR
Time To OCR 7 Images
FFTW
Build: Float + SSE - Size: 2D FFT Size 64
Git
Time To Complete Common Git Commands
FFTW
Build: Stock - Size: 2D FFT Size 1024
Radiance Benchmark
Test: SMP Parallel
FFTW
Build: Float + SSE - Size: 2D FFT Size 128
RAMspeed SMP
Type: Add - Benchmark: Floating Point
FFTW
Build: Float + SSE - Size: 1D FFT Size 64
PolyBench-C
Test: 3 Matrix Multiplications
Gzip Compression
Linux Source Tree Archiving To .tar.gz
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 32 - Solver Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 32
Radiance Benchmark
Test: Serial
GnuPG
Linux 4.3 Package File Encryption
LuaJIT
Test: Monte Carlo
Fhourstones
Complex Connect-4 Solving
Go Benchmarks
Test: build
Minion
Benchmark: Solitaire
RAMspeed SMP
Type: Add - Benchmark: Integer
PolyBench-C
Test: Covariance Computation
PolyBench-C
Test: Correlation Computation
FFTW
Build: Stock - Size: 1D FFT Size 256
GraphicsMagick
Operation: HWB Color Space
lzbench
Test: Zstd 1 - Process: Compression
FFTW
Build: Stock - Size: 2D FFT Size 128
lzbench
Test: XZ 0 - Process: Decompression
FFTW
Build: Stock - Size: 1D FFT Size 512
LuaJIT
Test: Jacobi Successive Over-Relaxation
FFTW
Build: Float + SSE - Size: 2D FFT Size 256
RAMspeed SMP
Type: Average - Benchmark: Floating Point
FFTW
Build: Float + SSE - Size: 2D FFT Size 512
Minion
Benchmark: Graceful
FFTW
Build: Stock - Size: 2D FFT Size 2048
FFTW
Build: Stock - Size: 1D FFT Size 2048
FFTW
Build: Stock - Size: 2D FFT Size 4096
Ogg Encoding
WAV To Ogg
FFTW
Build: Stock - Size: 1D FFT Size 1024
FFTW
Build: Stock - Size: 2D FFT Size 512
lzbench
Test: Zstd 1 - Process: Decompression
FFTW
Build: Stock - Size: 1D FFT Size 4096
GraphicsMagick
Operation: Rotate
FFTW
Build: Stock - Size: 2D FFT Size 256
RAMspeed SMP
Type: Copy - Benchmark: Floating Point
lzbench
Test: Brotli 0 - Process: Compression
RAMspeed SMP
Type: Scale - Benchmark: Floating Point
RAMspeed SMP
Type: Copy - Benchmark: Integer
RAMspeed SMP
Type: Triad - Benchmark: Floating Point
RAMspeed SMP
Type: Average - Benchmark: Integer
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 16 - Solver Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 16 - Solver Time
Cryptsetup
PBKDF2-whirlpool
Open Porous Media Git
OPM Benchmark: Flow MPI Extra - Threads: 32 - Solver Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 32 - Solver Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 64 - Solver Time
Open Porous Media Git
OPM Benchmark: Flow MPI Norne - Threads: 64
Sunflow Rendering System
Global Illumination + Image Synthesis
Memcached mcperf
Method: Replace
Memcached mcperf
Method: Prepend
Memcached mcperf
Method: Delete
Memcached mcperf
Method: Set
Memcached mcperf
Method: Get
Memcached mcperf
Method: Add
Stress-NG
Test: Context Switching
Timed MAFFT Alignment
Multiple Sequence Alignment
Rodinia
Test: OpenMP Streamcluster
Phoronix Test Suite v10.8.5