CPU Massive CPU Massive

A "massive" collection of primarily CPU benchmarks but also some memory (RAM) benchmarks for stressing modern processor hardware.

See how your system performs with this suite using the Phoronix Test Suite. It's as easy as running the phoronix-test-suite benchmark cpu-massive command..

Tests In This Suite

  • 7-Zip Compression

  • Apache Benchmark

  • Apache Siege

  •         Concurrent Users: 200
  •         Concurrent Users: 250
  •         Concurrent Users: 100

  • asmFish

  • BLAKE2

  • Blender

  •         Blend File: Fishy Cat - Compute: CPU-Only
  •         Blend File: Barbershop - Compute: CPU-Only
  •         Blend File: Pabellon Barcelona - Compute: CPU-Only
  •         Blend File: Classroom - Compute: CPU-Only
  •         Blend File: BMW27 - Compute: CPU-Only

  • BlogBench

  •         Test: Read
  •         Test: Write

  • Botan

  •         Test: Blowfish
  •         Test: CAST-256
  •         Test: KASUMI
  •         Test: Twofish
  •         Test: AES-256

  • BRL-CAD

  • C-Ray

  • CacheBench

  •         Test: Read
  •         Test: Write
  •         Test: Read / Modify / Write

  • Chaos Group V-RAY

  •         Mode: CPU

  • CLOMP

  • CloverLeaf

  • Compile Bench

  •         Test: Read Compiled Tree
  •         Test: Initial Create
  •         Test: Compile

  • CppPerformanceBenchmarks

  •         Test: Math Library
  •         Test: Function Objects
  •         Test: Atol
  •         Test: Ctype

  • Cpuminer-Opt

  •         Algorithm: sha256t
  •         Algorithm: skein
  •         Algorithm: m7m
  •         Algorithm: deep

  • Crafty

  • Cryptsetup

  • ctx_clock

  • Cython Benchmark

  •         Test: N-Queens

  • DaCapo Benchmark

  •         Java Test: Eclipse
  •         Java Test: H2
  •         Java Test: Jython
  •         Java Test: Tradebeans
  •         Java Test: Tradesoap

  • Darktable

  •         Test: Server Rack - Acceleration: CPU-only
  •         Test: Server Room - Acceleration: CPU-only
  •         Test: Masskrug - Acceleration: CPU-only
  •         Test: Boat - Acceleration: CPU-only

  • dav1d

  •         Video Input: Summer Nature 1080p
  •         Video Input: Summer Nature 4K

  • Dolfyn

  • ebizzy

  • FFTW

  •         Build: Float + SSE - Size: 2D FFT Size 4096
  •         Build: Float + SSE - Size: 1D FFT Size 4096
  •         Build: Stock - Size: 2D FFT Size 4096
  •         Build: Stock - Size: 1D FFT Size 4096

  • FLAC Audio Encoding

  • Geekbench

  •         Test: Multi Core
  •         Test: Single Core

  • glibc bench

  •         Benchmark: sqrt
  •         Benchmark: ffs
  •         Benchmark: pthread_once
  •         Benchmark: tanh
  •         Benchmark: sin
  •         Benchmark: ffsll
  •         Benchmark: cos

  • GNU Octave Benchmark

  • Go Benchmarks

  •         Test: build
  •         Test: http
  •         Test: json
  •         Test: garbage

  • GraphicsMagick

  •         Operation: Enhanced
  •         Operation: Resizing
  •         Operation: Rotate
  •         Operation: Sharpen

  • Hackbench

  •         Count: 32 - Type: Thread
  •         Count: 32 - Type: Process

  • Hierarchical INTegration

  •         Test: FLOAT
  •         Test: DOUBLE

  • High Performance Conjugate Gradient

  • Himeno Benchmark

  • HPC Challenge

  •         Test / Class: Random Ring Bandwidth
  •         Test / Class: Max Ping Pong Bandwidth
  •         Test / Class: Random Ring Latency
  •         Test / Class: EP-DGEMM
  •         Test / Class: EP-STREAM Triad
  •         Test / Class: G-Random Access
  •         Test / Class: G-Ptrans
  •         Test / Class: G-HPL
  •         Test / Class: G-Ffte

  • Java Gradle Build

  •         Gradle Build: Reactor

  • Java SciMark

  •         Computational Test: Dense LU Matrix Factorization
  •         Computational Test: Sparse Matrix Multiply
  •         Computational Test: Monte Carlo
  •         Computational Test: Jacobi Successive Over-Relaxation
  •         Computational Test: Fast Fourier Transform
  •         Computational Test: Composite

  • John The Ripper

  •         Test: Traditional DES
  •         Test: MD5
  •         Test: Blowfish

  • Kvazaar

  •         Video Input: Bosphorus 4K - Video Preset: Ultra Fast
  •         Video Input: Bosphorus 4K - Video Preset: Very Fast
  •         Video Input: Bosphorus 4K - Video Preset: Medium
  •         Video Input: Bosphorus 4K - Video Preset: Slow
  •         Video Input: Bosphorus 1080p - Video Preset: Ultra Fast
  •         Video Input: Bosphorus 1080p - Video Preset: Very Fast
  •         Video Input: Bosphorus 1080p - Video Preset: Medium
  •         Video Input: Bosphorus 1080p - Video Preset: Slow

  • LAME MP3 Encoding

  • LAMMPS Molecular Dynamics Simulator

  •         Test: Rhodopsin Protein

  • LeelaChessZero

  •         Backend: BLAS

  • libjpeg-turbo tjbench

  •         Test: Decompression Throughput

  • lzbench

  •         Test: Zstd 1
  •         Test: Brotli 0
  •         Test: Libdeflate 1
  •         Test: XZ 0

  • m-queens

  • MariaDB

  •         Clients: 16
  •         Clients: 1
  •         Clients: 256
  •         Clients: 64

  • MBW

  •         Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB
  •         Test: Memory Copy - Array Size: 128 MiB
  •         Test: Memory Copy - Array Size: 1024 MiB
  •         Test: Memory Copy - Array Size: 4096 MiB
  •         Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB
  •         Test: Memory Copy - Array Size: 8192 MiB
  •         Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB
  •         Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB

  • Memcached mcperf

  •         Method: Set
  •         Method: Replace
  •         Method: Delete
  •         Method: Prepend
  •         Method: Append
  •         Method: Add
  •         Method: Get

  • Minion

  •         Benchmark: Quasigroup
  •         Benchmark: Graceful
  •         Benchmark: Solitaire

  • Multichase Pointer Chaser

  •         Test: 256MB Array, 256 Byte Stride
  •         Test: 4MB Array, 64 Byte Stride
  •         Test: 1GB Array, 256 Byte Stride, 4 Threads
  •         Test: 1GB Array, 256 Byte Stride, 2 Threads
  •         Test: 1GB Array, 256 Byte Stride

  • NAMD

  • NAS Parallel Benchmarks

  •         Test / Class: LU.C
  •         Test / Class: LU.A
  •         Test / Class: FT.B
  •         Test / Class: FT.A
  •         Test / Class: EP.C
  •         Test / Class: BT.A
  •         Test / Class: SP.A

  • NGINX Benchmark

  • Node.js Express HTTP Load Test

  • Novabench

  •         Test: RAM
  •         Test: CPU

  • Numenta Anomaly Benchmark

  •         Detector: EXPoSE
  •         Detector: Earthgecko Skyline
  •         Detector: Relative Entropy
  •         Detector: Windowed Gaussian
  •         Detector: Bayesian Changepoint

  • Numpy Benchmark

  • oneDNN

  •         Harness: Convolution Batch conv_alexnet - Data Type: f32
  •         Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32
  •         Harness: Deconvolution Batch deconv_1d - Data Type: f32
  •         Harness: Deconvolution Batch deconv_3d - Data Type: f32
  •         Harness: IP Batch 1D - Data Type: f32

  • Open FMM Nero2D

  • OpenSSL

  • Parallel BZIP2 Compression

  • Parboil

  •         Test: OpenMP Stencil
  •         Test: OpenMP CUTCP
  •         Test: OpenMP MRI-Q
  •         Test: OpenMP MRI Gridding
  •         Test: OpenMP LBM

  • PHPBench

  • PlaidML

  •         FP16: No - Mode: Inference - Network: VGG16 - Device: CPU
  •         FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
  •         FP16: No - Mode: Training - Network: VGG16 - Device: CPU
  •         FP16: No - Mode: Training - Network: ResNet 50 - Device: CPU

  • PostgreSQL pgbench

  •         Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
  •         Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only
  •         Scaling: Buffer Test - Test: Normal Load - Mode: Read Write
  •         Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write

  • POV-Ray

  • Primesieve

  • PyMongo Inserts

  • R Benchmark

  • Radiance Benchmark

  •         Test: Serial
  •         Test: SMP Parallel

  • RAMspeed SMP

  •         Type: Triad - Benchmark: Integer
  •         Type: Add - Benchmark: Floating Point
  •         Type: Scale - Benchmark: Integer
  •         Type: Add - Benchmark: Integer
  •         Type: Copy - Benchmark: Floating Point
  •         Type: Copy - Benchmark: Integer
  •         Type: Average - Benchmark: Floating Point
  •         Type: Scale - Benchmark: Floating Point
  •         Type: Triad - Benchmark: Floating Point
  •         Type: Average - Benchmark: Integer

  • Redis

  •         Test: LPOP
  •         Test: LPUSH
  •         Test: SADD
  •         Test: SET
  •         Test: GET

  • Renaissance

  •         Test: Scala Dotty
  •         Test: In-Memory Database Shootout
  •         Test: Apache Spark PageRank
  •         Test: Apache Spark Bayes
  •         Test: Twitter Finagle
  •         Test: Akka Unbalanced Cobwebbed Tree
  •         Test: Savina Reactors.IO
  •         Test: Apache Spark ALS

  • Rodinia

  •         Test: OpenMP LavaMD
  •         Test: OpenMP Streamcluster
  •         Test: OpenMP Leukocyte
  •         Test: OpenMP CFD Solver

  • Rust Mandelbrot

  • Rust Prime Benchmark

  • Scikit-Learn

  • Sockperf

  •         Test: Latency Ping Pong
  •         Test: Throughput
  •         Test: Latency Under Load

  • SPECjbb 2015

  • Stockfish

  • Stream

  •         Type: Scale
  •         Type: Add
  •         Type: Triad
  •         Type: Copy

  • Stress-NG

  •         Test: Context Switching
  •         Test: Socket Activity
  •         Test: Semaphores
  •         Test: System V Message Passing
  •         Test: Forking
  •         Test: Memory Copying
  •         Test: Crypto
  •         Test: CPU Stress

  • SVT-AV1

  •         Encoder Mode: Enc Mode 8 - Input: 1080p
  •         Encoder Mode: Enc Mode 4 - Input: 1080p
  •         Encoder Mode: Enc Mode 0 - Input: 1080p

  • SVT-HEVC

  • SVT-VP9

  •         Tuning: VMAF Optimized - Input: Bosphorus 1080p
  •         Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p
  •         Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

  • Swet

  • Sysbench

  •         Test: Memory
  •         Test: CPU

  • t-test1

  •         Threads: 2
  •         Threads: 1

  • Tachyon

  • Tensorflow

  •         Build: Cifar10

  • Tesseract OCR

  • Timed Apache Compilation

  • Timed GCC Compilation

  • Timed HMMer Search

  • Timed Linux Kernel Compilation

  • Timed LLVM Compilation

  • Timed MAFFT Alignment

  • Timed MrBayes Analysis

  • Timed PHP Compilation

  • Tinymembench

  • TTSIOD 3D Renderer

  • VP9 libvpx Encoding

  •         Speed: Speed 5
  •         Speed: Speed 0

  • x264

  • x265

  •         Video Input: Bosphorus 1080p
  •         Video Input: Bosphorus 4K

  • Xsbench

  • XZ Compression

  • Zstd Compression

  •         Compression Level: 3, Long Mode
  •         Compression Level: 8
  •         Compression Level: 8, Long Mode
  •         Compression Level: 19
  •         Compression Level: 19, Long Mode
  •         Compression Level: 3

Revision History Revision History

pts/cpu-massive-1.0.4     Fri, 16 Oct 2020 18:38:00 GMT
Update x265, add kvazaar, mark as deprecated.

pts/cpu-massive-1.0.3     Wed, 17 Jun 2020 16:34:18 GMT
Drop version bind, rename mkl-dnn test profile to onednn

pts/cpu-massive-1.0.2     Wed, 08 Apr 2020 16:03:35 GMT
Mark suite as deprecated since quite big and easier to manage with virtual suite.

pts/cpu-massive-1.0.1     Fri, 14 Jun 2019 08:26:48 GMT
Drop "mostly RAM" pgbench test cases since on large servers it can fill up the disk from taking up too much space.

pts/cpu-massive-1.0.0     Thu, 13 Jun 2019 20:33:21 GMT
Initial commit.