Extra EPYC Rome Benchmarks
2 x Intel Xeon Platinum 8280 testing with a GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS) and llvmpipe 377GB on Ubuntu 19.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1908130-AS-EXTRAEPYC84&grs.
MKL-DNN
Harness: Deconvolution Batch deconv_1d - Data Type: f32
MKL-DNN
Harness: Convolution Batch conv_alexnet - Data Type: u8s8u8s32
MKL-DNN
Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32s32
Parboil
Test: OpenMP MRI Gridding
MKL-DNN
Harness: Deconvolution Batch deconv_all - Data Type: u8s8u8s32
MKL-DNN
Harness: Deconvolution Batch deconv_1d - Data Type: u8s8u8s32
MKL-DNN
Harness: IP Batch All - Data Type: u8s8f32s32
MKL-DNN
Harness: IP Batch All - Data Type: u8s8u8s32
MKL-DNN
Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32s32
MKL-DNN
Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8u8s32
MKL-DNN
Harness: Deconvolution Batch deconv_3d - Data Type: f32
MBW
Test: Memory Copy - Array Size: 8192 MiB
MBW
Test: Memory Copy - Array Size: 1024 MiB
MBW
Test: Memory Copy - Array Size: 4096 MiB
MBW
Test: Memory Copy - Array Size: 512 MiB
MKL-DNN
Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32s32
CP2K Molecular Dynamics
Fayalite-FIST Data
Himeno Benchmark
Poisson Pressure Solver
Memcached mcperf
Method: Replace
Memcached mcperf
Method: Prepend
Memcached mcperf
Method: Get
Memcached mcperf
Method: Append
MKL-DNN
Harness: IP Batch 1D - Data Type: u8s8f32s32
MKL-DNN
Harness: IP Batch 1D - Data Type: u8s8u8s32
MKL-DNN
Harness: Convolution Batch conv_all - Data Type: u8s8u8s32
MKL-DNN
Harness: Convolution Batch conv_all - Data Type: u8s8f32s32
HPC Challenge
Test / Class: G-Ffte
HPC Challenge
Test / Class: G-Ffte
BLAKE2
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB
Stress-NG
Test: System V Message Passing
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB
Memcached mcperf
Method: Add
Memcached mcperf
Method: Set
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB
MBW
Test: Memory Copy - Array Size: 128 MiB
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB
Apache Siege
Concurrent Users: 1
MariaDB
Clients: 4
BlogBench
Test: Read
RAMspeed SMP
Type: Add - Benchmark: Floating Point
MariaDB
Clients: 16
MKL-DNN
Harness: Convolution Batch conv_alexnet - Data Type: f32
MariaDB
Clients: 32
HPC Challenge
Test / Class: Random Ring Bandwidth
OSBench
Test: Create Threads
RAMspeed SMP
Type: Triad - Benchmark: Floating Point
MariaDB
Clients: 8
MKL-DNN
Harness: Convolution Batch conv_all - Data Type: f32
MariaDB
Clients: 256
dav1d
Video Input: Summer Nature 1080p
Parboil
Test: OpenMP CUTCP
MKL-DNN
Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32
Stress-NG
Test: Tsearch
Stress-NG
Test: Forking
HPC Challenge
Test / Class: Max Ping Pong Bandwidth
MariaDB
Clients: 128
DaCapo Benchmark
Java Test: Tradebeans
Stress-NG
Test: Socket Activity
HPC Challenge
Test / Class: Random Ring Latency
RAMspeed SMP
Type: Scale - Benchmark: Floating Point
SVT-AV1
1080p 8-bit YUV To AV1 Video Encode
MariaDB
Clients: 64
Appleseed
Scene: Material Tester
FFTW
Build: Float + SSE - Size: 2D FFT Size 512
VP9 libvpx Encoding
vpxenc VP9 1080p Video Encode
RAMspeed SMP
Type: Average - Benchmark: Floating Point
Bork File Encrypter
File Encryption Time
FFTW
Build: Float + SSE - Size: 2D FFT Size 1024
FFTW
Build: Float + SSE - Size: 1D FFT Size 1024
Stress-NG
Test: Semaphores
FFTW
Build: Float + SSE - Size: 1D FFT Size 512
RAMspeed SMP
Type: Add - Benchmark: Integer
Tinymembench
Standard Memset
MKL-DNN
Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
MKL-DNN
Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32
OSBench
Test: Launch Programs
NAS Parallel Benchmarks
Test / Class: FT.B
FFTW
Build: Float + SSE - Size: 1D FFT Size 2048
NAS Parallel Benchmarks
Test / Class: FT.A
BlogBench
Test: Write
FFTW
Build: Float + SSE - Size: 1D FFT Size 4096
FFTW
Build: Float + SSE - Size: 2D FFT Size 256
OSBench
Test: Create Processes
dav1d
Video Input: Summer Nature 4K
Java SciMark
Computational Test: Monte Carlo
SVT-VP9
1080p 8-bit YUV To VP9 Video Encode
MKL-DNN
Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32s32
Numpy Benchmark
Stress-NG
Test: Crypto
MKL-DNN
Harness: Deconvolution Batch deconv_3d - Data Type: u8s8u8s32
RAMspeed SMP
Type: Triad - Benchmark: Integer
m-queens
Time To Solve
RAMspeed SMP
Type: Average - Benchmark: Integer
Sockperf
Test: Latency Ping Pong
x265
H.265 1080p Video Encoding
RAMspeed SMP
Type: Copy - Benchmark: Integer
RAMspeed SMP
Type: Scale - Benchmark: Integer
DaCapo Benchmark
Java Test: Tradesoap
C-Ray
Total Time - 4K, 16 Rays Per Pixel
Tinymembench
Standard Memcpy
SVT-HEVC
1080p 8-bit YUV To HEVC Video Encode
John The Ripper
Test: Blowfish
PHPBench
PHP Benchmark Suite
Apache Siege
Concurrent Users: 10
PHP Micro Benchmarks
Test: Zend micro_bench
FFTW
Build: Float + SSE - Size: 2D FFT Size 32
Stress-NG
Test: Hsearch
Apache Siege
Concurrent Users: 50
lzbench
Test: XZ 0 - Process: Compression
IndigoBench
Scene: Bedroom
Phoronix Test Suite Self Test
Test: Info Command
IndigoBench
Scene: Supercar
DaCapo Benchmark
Java Test: H2
Renaissance
Test: In-Memory Database Shootout
FFTW
Build: Float + SSE - Size: 2D FFT Size 2048
Cython benchmark
Apache Siege
Concurrent Users: 100
Rodinia
Test: OpenMP CFD Solver
Zstd Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19
x264
H.264 Video Encoding
Minion
Benchmark: Quasigroup
Phoronix Test Suite Self Test
Test: Merge Results
dcraw
RAW To PPM Image Conversion
MKL-DNN
Harness: IP Batch All - Data Type: f32
Darktable
Test: Boat - Acceleration: CPU-only
FFTW
Build: Float + SSE - Size: 2D FFT Size 128
Botan
Test: Blowfish - Encrypt
MKL-DNN
Harness: IP Batch 1D - Data Type: f32
TSCP
AI Chess Performance
Botan
Test: AES-256 - Decrypt
FFTW
Build: Float + SSE - Size: 1D FFT Size 32
lzbench
Test: Libdeflate 1 - Process: Decompression
Botan
Test: Blowfish - Decrypt
PHP Micro Benchmarks
Test: Zend bench
Botan
Test: AES-256 - Encrypt
XZ Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
Timed GCC Compilation
Time To Compile
SPECjbb 2015
SPECjbb2015-Composite critical-jOPS
Stream
Type: Scale
Stress-NG
Test: Glibc Qsort Data Sorting
DaCapo Benchmark
Java Test: Jython
Redis
Test: SADD
FFTW
Build: Stock - Size: 2D FFT Size 2048
lzbench
Test: Brotli 0 - Process: Decompression
Phoronix Test Suite Self Test
Test: Total Time
RAMspeed SMP
Type: Copy - Benchmark: Floating Point
Botan
Test: Twofish - Decrypt
Apache Siege
Concurrent Users: 200
Blender
Blend File: Classroom - Compute: CPU-Only
Java SciMark
Computational Test: Fast Fourier Transform
Phoronix Test Suite Self Test
Test: Render Test
Botan
Test: Twofish - Encrypt
HPC Challenge
Test / Class: G-HPL
Botan
Test: KASUMI - Decrypt
HPC Challenge
Test / Class: G-Random Access
Apache Siege
Concurrent Users: 250
Botan
Test: KASUMI - Encrypt
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
Stress-NG
Test: Memory Copying
Mencoder
AVI To LAVC
Botan
Test: CAST-256 - Decrypt
Botan
Test: CAST-256 - Encrypt
SPECjbb 2015
SPECjbb2015-Composite max-jOPS
GNU GMP GMPbench
Total Time
Darktable
Test: Masskrug - Acceleration: CPU-only
Appleseed
Scene: Disney Material
HPC Challenge
Test / Class: EP-DGEMM
FFTW
Build: Stock - Size: 2D FFT Size 4096
Primesieve
1e12 Prime Number Generation
POV-Ray
Trace Time
lzbench
Test: Libdeflate 1 - Process: Compression
Timed LLVM Compilation
Time To Compile
Java SciMark
Computational Test: Dense LU Matrix Factorization
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
libjpeg-turbo tjbench
Test: Decompression Throughput
Ogg Encoding
WAV To Ogg
Appleseed
Scene: Emily
Coremark
CoreMark Size 666 - Iterations Per Second
Java SciMark
Computational Test: Composite
FFTW
Build: Float + SSE - Size: 2D FFT Size 4096
FFTW
Build: Stock - Size: 1D FFT Size 64
LAME MP3 Encoding
WAV To MP3
OSBench
Test: Create Files
FFTW
Build: Stock - Size: 1D FFT Size 512
OSBench
Test: Memory Allocations
ctx_clock
Context Switch Time
Blender
Blend File: Classroom - Compute: CPU-Only
FFTW
Build: Stock - Size: 2D FFT Size 128
FFTW
Build: Float + SSE - Size: 2D FFT Size 64
7-Zip Compression
Compress Speed Test
Blender
Blend File: Fishy Cat - Compute: CPU-Only
Stress-NG
Test: Glibc C String Functions
lzbench
Test: XZ 0 - Process: Decompression
Blender
Blend File: Fishy Cat - Compute: CPU-Only
GIMP
Test: rotate
NAS Parallel Benchmarks
Test / Class: EP.C
Timed ImageMagick Compilation
Time To Compile
GIMP
Test: unsharp-mask
FFTW
Build: Stock - Size: 1D FFT Size 4096
Blender
Blend File: BMW27 - Compute: CPU-Only
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Write
GIMP
Test: auto-levels
Stress-NG
Test: Vector Math
FFTW
Build: Stock - Size: 1D FFT Size 2048
Stockfish
Total Time
PostgreSQL pgbench
Scaling: Buffer Test - Test: Single Thread - Mode: Read Only
HPC Challenge
Test / Class: EP-STREAM Triad
GIMP
Test: resize
Tesseract OCR
Time To OCR 7 Images
Stress-NG
Test: Lsearch
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
Stress-NG
Test: CPU Stress
FFTW
Build: Stock - Size: 1D FFT Size 1024
lzbench
Test: Zstd 1 - Process: Compression
Stress-NG
Test: Bsearch
Renaissance
Test: Scala Dotty
FFTW
Build: Stock - Size: 2D FFT Size 512
FFTW
Build: Stock - Size: 2D FFT Size 256
FLAC Audio Encoding
WAV To FLAC
Java SciMark
Computational Test: Jacobi Successive Over-Relaxation
Chaos Group V-RAY
Mode: CPU
Blender
Blend File: BMW27 - Compute: CPU-Only
NAS Parallel Benchmarks
Test / Class: LU.C
OpenSSL
RSA 4096-bit Performance
NAS Parallel Benchmarks
Test / Class: SP.A
Phoronix Test Suite Self Test
Test: List Suites
Java SciMark
Computational Test: Sparse Matrix Multiply
FFTW
Build: Stock - Size: 2D FFT Size 1024
FFTW
Build: Stock - Size: 1D FFT Size 128
lzbench
Test: Zstd 1 - Process: Decompression
FFTW
Build: Float + SSE - Size: 1D FFT Size 128
FFTW
Build: Stock - Size: 2D FFT Size 64
Minion
Benchmark: Solitaire
Y-Cruncher
Calculating 500M Pi Digits
lzbench
Test: Brotli 0 - Process: Compression
Blender
Blend File: Barbershop - Compute: CPU-Only
FFTW
Build: Stock - Size: 2D FFT Size 32
Stress-NG
Test: Matrix Math
FFTW
Build: Stock - Size: 1D FFT Size 256
Blender
Blend File: Barbershop - Compute: CPU-Only
PostgreSQL pgbench
Scaling: Buffer Test - Test: Single Thread - Mode: Read Write
Timed PHP Compilation
Time To Compile
Minion
Benchmark: Graceful
asmFish
1024 Hash Memory, 26 Depth
NAS Parallel Benchmarks
Test / Class: BT.A
Phoronix Test Suite Self Test
Test: Peak Memory Usage
FFTE
Test: N=256, 1D Complex FFT Routine
Sunflow Rendering System
Global Illumination + Image Synthesis
Memcached mcperf
Method: Delete
Stress-NG
Test: Context Switching
Redis
Test: SET
Redis
Test: GET
Redis
Test: LPUSH
Redis
Test: LPOP
Darktable
Test: Server Room - Acceleration: CPU-only
Darktable
Test: Server Rack - Acceleration: CPU-only
Timed Linux Kernel Compilation
Time To Compile
MKL-DNN
Harness: Deconvolution Batch deconv_all - Data Type: f32
MKL-DNN
Harness: Convolution Batch conv_3d - Data Type: f32
Renaissance
Test: Akka Unbalanced Cobwebbed Tree
Renaissance
Test: Apache Spark PageRank
Renaissance
Test: Savina Reactors.IO
Renaissance
Test: Apache Spark Bayes
Renaissance
Test: Apache Spark ALS
Timed MAFFT Alignment
Multiple Sequence Alignment
FFTW
Build: Float + SSE - Size: 1D FFT Size 256
FFTW
Build: Float + SSE - Size: 1D FFT Size 64
FFTW
Build: Stock - Size: 1D FFT Size 32
NAMD
ATPase Simulation - 327,506 Atoms
CLOMP
Static OMP Speedup
High Performance Conjugate Gradient
Rodinia
Test: OpenMP Streamcluster
Rodinia
Test: OpenMP LavaMD
Parboil
Test: OpenMP Stencil
Parboil
Test: OpenMP LBM
HPC Challenge
Test / Class: G-Ptrans
NAS Parallel Benchmarks
Test / Class: LU.A
Sockperf
Test: Latency Under Load
Sockperf
Test: Throughput
Stream
Type: Add
Stream
Type: Triad
Stream
Type: Copy
Phoronix Test Suite v10.8.4