Linux CPU Benchmarks
Tests for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2301170-PTS-13400ALO23&grs.
Stress-NG
Test: NUMA
NAS Parallel Benchmarks
Test / Class: IS.D
Intel Open Image Denoise
Run: RT.ldr_alb_nrm.3840x2160
Intel Open Image Denoise
Run: RT.hdr_alb_nrm.3840x2160
Redis
Test: GET - Parallel Connections: 500
LuaRadio
Test: Five Back to Back FIR Filters
Nettle
Test: sha512
C-Blosc
Test: blosclz bitshuffle
NCNN
Target: CPU - Model: googlenet
FinanceBench
Benchmark: Repo OpenMP
WireGuard + Linux Networking Stack Stress Test
Natron
Input: Spaceship
GROMACS
Implementation: MPI CPU - Input: water_GMX50_bare
C-Blosc
Test: blosclz shuffle
GIMP
Test: resize
GNU Radio
Test: Five Back to Back FIR Filters
Selenium
Benchmark: Jetstream 2 - Browser: Firefox
Redis
Test: SET - Parallel Connections: 50
Stargate Digital Audio Workstation
Sample Rate: 480000 - Buffer Size: 512
GIMP
Test: rotate
Rodinia
Test: OpenMP Leukocyte
NAS Parallel Benchmarks
Test / Class: FT.C
GIMP
Test: unsharp-mask
GIMP
Test: auto-levels
Timed MrBayes Analysis
Primate Phylogeny Analysis
DaCapo Benchmark
Java Test: H2
x265
Video Input: Bosphorus 1080p
NCNN
Target: CPU - Model: yolov4-tiny
Zstd Compression
Compression Level: 19 - Compression Speed
ClickHouse
100M Rows Web Analytics Dataset, First Run / Cold Cache
Stargate Digital Audio Workstation
Sample Rate: 96000 - Buffer Size: 512
Hierarchical INTegration
Test: FLOAT
PyPerformance
Benchmark: python_startup
x265
Video Input: Bosphorus 4K
GraphicsMagick
Operation: HWB Color Space
Stargate Digital Audio Workstation
Sample Rate: 44100 - Buffer Size: 512
Node.js V8 Web Tooling Benchmark
ctx_clock
Context Switch Time
NCNN
Target: CPU - Model: vision_transformer
Selenium
Benchmark: Speedometer - Browser: Firefox
Node.js Express HTTP Load Test
Pennant
Test: leblancbig
RawTherapee
Total Benchmark Time
Stargate Digital Audio Workstation
Sample Rate: 96000 - Buffer Size: 1024
ClickHouse
100M Rows Web Analytics Dataset, Third Run
Inkscape
Operation: SVG Files To PNG
Timed HMMer Search
Pfam Database Search
Liquid-DSP
Threads: 1 - Buffer Length: 256 - Filter Length: 57
Xcompact3d Incompact3d
Input: input.i3d 129 Cells Per Direction
Zstd Compression
Compression Level: 19, Long Mode - Compression Speed
ClickHouse
100M Rows Web Analytics Dataset, Second Run
LuxCoreRender
Scene: DLSC - Acceleration: CPU
ASKAP
Test: tConvolve OpenMP - Degridding
Xmrig
Variant: Wownero - Hash Count: 1M
GPAW
Input: Carbon Nanotube
Selenium
Benchmark: Kraken - Browser: Firefox
GEGL
Operation: Reflect
Timed MPlayer Compilation
Time To Compile
Algebraic Multi-Grid Benchmark
Selenium
Benchmark: ARES-6 - Browser: Firefox
GraphicsMagick
Operation: Resizing
DaCapo Benchmark
Java Test: Tradebeans
LibRaw
Post-Processing Benchmark
Selenium
Benchmark: WASM collisionDetection - Browser: Google Chrome
Timed Linux Kernel Compilation
Build: defconfig
asmFish
1024 Hash Memory, 26 Depth
Timed LLVM Compilation
Build System: Unix Makefiles
PyPerformance
Benchmark: nbody
WebP2 Image Encode
Encode Settings: Quality 100, Compression Effort 5
NCNN
Target: CPU - Model: vgg16
Xcompact3d Incompact3d
Input: input.i3d 193 Cells Per Direction
Timed Linux Kernel Compilation
Build: allmodconfig
LuxCoreRender
Scene: Danish Mood - Acceleration: CPU
Selenium
Benchmark: Kraken - Browser: Google Chrome
High Performance Conjugate Gradient
GraphicsMagick
Operation: Enhanced
Unpacking Firefox
Extracting: firefox-84.0.source.tar.xz
Stargate Digital Audio Workstation
Sample Rate: 480000 - Buffer Size: 1024
Selenium
Benchmark: ARES-6 - Browser: Google Chrome
NCNN
Target: CPU - Model: alexnet
SVT-AV1
Encoder Mode: Preset 8 - Input: Bosphorus 4K
CloverLeaf
Lagrangian-Eulerian Hydrodynamics
TensorFlow Lite
Model: Mobilenet Float
AI Benchmark Alpha
Device Inference Score
Cryptsetup
AES-XTS 256b Decryption
LZ4 Compression
Compression Level: 9 - Decompression Speed
Embree
Binary: Pathtracer - Model: Asian Dragon
libavif avifenc
Encoder Speed: 0
Liquid-DSP
Threads: 2 - Buffer Length: 256 - Filter Length: 57
Timed Godot Game Engine Compilation
Time To Compile
Stargate Digital Audio Workstation
Sample Rate: 44100 - Buffer Size: 1024
Embree
Binary: Pathtracer ISPC - Model: Crown
SVT-AV1
Encoder Mode: Preset 10 - Input: Bosphorus 4K
Timed Mesa Compilation
Time To Compile
SVT-VP9
Tuning: Visual Quality Optimized - Input: Bosphorus 4K
Selenium
Benchmark: PSPDFKit WASM - Browser: Google Chrome
Liquid-DSP
Threads: 8 - Buffer Length: 256 - Filter Length: 57
AI Benchmark Alpha
Device AI Score
Ngspice
Circuit: C2670
AI Benchmark Alpha
Device Training Score
WebP2 Image Encode
Encode Settings: Default
Timed FFmpeg Compilation
Time To Compile
SVT-HEVC
Tuning: 10 - Input: Bosphorus 4K
DaCapo Benchmark
Java Test: Jython
TNN
Target: CPU - Model: DenseNet
PyPerformance
Benchmark: 2to3
Nebular Empirical Analysis Tool
OSPRay Studio
Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
WebP2 Image Encode
Encode Settings: Quality 100, Lossless Compression
NAS Parallel Benchmarks
Test / Class: LU.C
G'MIC
Test: 2D Function Plotting, 1000 Times
GNU Radio
Test: IIR Filter
GEGL
Operation: Antialias
Zstd Compression
Compression Level: 8 - Compression Speed
Coremark
CoreMark Size 666 - Iterations Per Second
AOM AV1
Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K
GNU Radio
Test: Hilbert Transform
SVT-AV1
Encoder Mode: Preset 4 - Input: Bosphorus 1080p
Selenium
Benchmark: Jetstream 2 - Browser: Google Chrome
VP9 libvpx Encoding
Speed: Speed 5 - Input: Bosphorus 4K
Cryptsetup
AES-XTS 512b Encryption
OSPRay
Benchmark: particle_volume/scivis/real_time
NAS Parallel Benchmarks
Test / Class: EP.C
Liquid-DSP
Threads: 4 - Buffer Length: 256 - Filter Length: 57
PyPerformance
Benchmark: raytrace
Liquid-DSP
Threads: 32 - Buffer Length: 256 - Filter Length: 57
LZ4 Compression
Compression Level: 1 - Decompression Speed
Rodinia
Test: OpenMP CFD Solver
Chaos Group V-RAY
Mode: CPU
NCNN
Target: CPU - Model: resnet50
SVT-VP9
Tuning: VMAF Optimized - Input: Bosphorus 1080p
OpenSSL
Algorithm: SHA256
LeelaChessZero
Backend: BLAS
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
Selenium
Benchmark: PSPDFKit WASM - Browser: Firefox
SVT-AV1
Encoder Mode: Preset 10 - Input: Bosphorus 1080p
SVT-AV1
Encoder Mode: Preset 12 - Input: Bosphorus 1080p
SVT-VP9
Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p
TNN
Target: CPU - Model: SqueezeNet v2
Etcpak
Benchmark: Multi-Threaded - Configuration: ETC2
SVT-HEVC
Tuning: 7 - Input: Bosphorus 1080p
ASKAP
Test: tConvolve OpenMP - Gridding
TensorFlow Lite
Model: Mobilenet Quant
NCNN
Target: CPU - Model: resnet18
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Very Fast
libgav1
Video Input: Summer Nature 4K
GEGL
Operation: Color Enhance
LAMMPS Molecular Dynamics Simulator
Model: 20k Atoms
Timed Gem5 Compilation
Time To Compile
LuaRadio
Test: Complex Phase
Timed LLVM Compilation
Build System: Ninja
LuxCoreRender
Scene: Orange Juice - Acceleration: CPU
NAS Parallel Benchmarks
Test / Class: BT.C
x264
Video Input: Bosphorus 1080p
GNU Radio
Test: FM Deemphasis Filter
TensorFlow Lite
Model: SqueezeNet
Appleseed
Scene: Disney Material
OSPRay
Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time
Rodinia
Test: OpenMP LavaMD
Cryptsetup
AES-XTS 512b Decryption
SVT-VP9
Tuning: VMAF Optimized - Input: Bosphorus 4K
OSPRay
Benchmark: gravity_spheres_volume/dim_512/scivis/real_time
Selenium
Benchmark: WASM imageConvolute - Browser: Google Chrome
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Very Fast
SVT-VP9
Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
GNU Octave Benchmark
libavif avifenc
Encoder Speed: 6
Etcpak
Benchmark: Single-Threaded - Configuration: ETC2
ASKAP
Test: tConvolve MT - Degridding
ASKAP
Test: Hogbom Clean OpenMP
LuxCoreRender
Scene: Rainbow Colors and Prism - Acceleration: CPU
AOM AV1
Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K
Unpacking The Linux Kernel
linux-5.19.tar.xz
Google Draco
Model: Church Facade
simdjson
Throughput Test: Kostya
NAS Parallel Benchmarks
Test / Class: CG.C
Parallel BZIP2 Compression
FreeBSD-13.0-RELEASE-amd64-memstick.img Compression
x264
Video Input: Bosphorus 4K
OSPRay
Benchmark: particle_volume/pathtracer/real_time
OpenSCAD
Render: Mini-ITX Case
SVT-VP9
Tuning: Visual Quality Optimized - Input: Bosphorus 1080p
TNN
Target: CPU - Model: MobileNet v2
Embree
Binary: Pathtracer - Model: Crown
OSPRay Studio
Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
Crafty
Elapsed Time
SQLite Speedtest
Timed Time - Size 1,000
OSPRay
Benchmark: particle_volume/ao/real_time
AOM AV1
Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K
PyBench
Total For Average Test Times
OSPRay
Benchmark: gravity_spheres_volume/dim_512/ao/real_time
Cryptsetup
Serpent-XTS 512b Encryption
Cryptsetup
Twofish-XTS 256b Decryption
Cryptsetup
Serpent-XTS 256b Encryption
simdjson
Throughput Test: DistinctUserID
NAS Parallel Benchmarks
Test / Class: SP.B
GEGL
Operation: Wavelet Blur
LZ4 Compression
Compression Level: 3 - Decompression Speed
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
m-queens
Time To Solve
GNU Radio
Test: Signal Source (Cosine)
Blender
Blend File: BMW27 - Compute: CPU-Only
WavPack Audio Encoding
WAV To WavPack
WebP2 Image Encode
Encode Settings: Quality 75, Compression Effort 7
LZ4 Compression
Compression Level: 9 - Compression Speed
G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
Cryptsetup
Twofish-XTS 512b Decryption
Cryptsetup
Serpent-XTS 512b Decryption
LuaRadio
Test: FM Deemphasis Filter
OSPRay Studio
Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
OpenSCAD
Render: Pistol
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Ultra Fast
Xmrig
Variant: Monero - Hash Count: 1M
Cryptsetup
Serpent-XTS 256b Decryption
Gzip Compression
Linux Source Tree Archiving To .tar.gz
Selenium
Benchmark: WASM collisionDetection - Browser: Firefox
SVT-AV1
Encoder Mode: Preset 4 - Input: Bosphorus 4K
Zstd Compression
Compression Level: 19, Long Mode - Decompression Speed
Appleseed
Scene: Emily
ASKAP
Test: tConvolve MT - Gridding
Blender
Blend File: Classroom - Compute: CPU-Only
PyPerformance
Benchmark: regex_compile
TSCP
AI Chess Performance
OpenSSL
Algorithm: RSA4096
libjpeg-turbo tjbench
Test: Decompression Throughput
Appleseed
Scene: Material Tester
Ngspice
Circuit: C7552
Cryptsetup
Twofish-XTS 256b Encryption
NAS Parallel Benchmarks
Test / Class: MG.C
LZ4 Compression
Compression Level: 1 - Compression Speed
Zstd Compression
Compression Level: 19 - Decompression Speed
Chia Blockchain VDF
Test: Square Plain C++
Google Draco
Model: Lion
libavif avifenc
Encoder Speed: 10, Lossless
Zstd Compression
Compression Level: 8 - Decompression Speed
SVT-AV1
Encoder Mode: Preset 12 - Input: Bosphorus 4K
Blender
Blend File: Fishy Cat - Compute: CPU-Only
OpenSCAD
Render: Leonardo Phone Case Slim
Cryptsetup
Twofish-XTS 512b Encryption
oneDNN
Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
GNU Radio
Test: FIR Filter
WebP2 Image Encode
Encode Settings: Quality 95, Compression Effort 7
LeelaChessZero
Backend: Eigen
Liquid-DSP
Threads: 16 - Buffer Length: 256 - Filter Length: 57
libavif avifenc
Encoder Speed: 6, Lossless
N-Queens
Elapsed Time
SecureMark
Benchmark: SecureMark-TLS
IndigoBench
Acceleration: CPU - Scene: Bedroom
OSPRay Studio
Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer
Primesieve
Length: 1e12
PyHPC Benchmarks
Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing
LuaRadio
Test: Hilbert Transform
OSPRay Studio
Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer
TensorFlow Lite
Model: Inception V4
LZ4 Compression
Compression Level: 3 - Compression Speed
OSPRay Studio
Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
Nettle
Test: aes256
Zstd Compression
Compression Level: 8, Long Mode - Decompression Speed
NAMD
ATPase Simulation - 327,506 Atoms
Cryptsetup
PBKDF2-sha512
Helsing
Digit Range: 14 digit
TNN
Target: CPU - Model: SqueezeNet v1.1
Cryptsetup
PBKDF2-whirlpool
POV-Ray
Trace Time
Zstd Compression
Compression Level: 8, Long Mode - Compression Speed
libavif avifenc
Encoder Speed: 2
ASTC Encoder
Preset: Medium
Zstd Compression
Compression Level: 3, Long Mode - Decompression Speed
Blender
Blend File: Barbershop - Compute: CPU-Only
oneDNN
Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU
oneDNN
Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU
IndigoBench
Acceleration: CPU - Scene: Supercar
Nettle
Test: poly1305-aes
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
QuantLib
Bork File Encrypter
File Encryption Time
oneDNN
Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU
OSPRay Studio
Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
SVT-HEVC
Tuning: 10 - Input: Bosphorus 1080p
Zstd Compression
Compression Level: 3, Long Mode - Compression Speed
Nettle
Test: chacha
OCRMyPDF
Processing 60 Page PDF Document
OSPRay Studio
Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer
Sysbench
Test: CPU
Aircrack-ng
eSpeak-NG Speech Engine
Text-To-Speech Synthesis
ASTC Encoder
Preset: Thorough
SVT-AV1
Encoder Mode: Preset 8 - Input: Bosphorus 1080p
OSPRay Studio
Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
OpenSSL
Algorithm: RSA4096
SVT-HEVC
Tuning: 7 - Input: Bosphorus 4K
ASTC Encoder
Preset: Exhaustive
GraphicsMagick
Operation: Noise-Gaussian
OpenVKL
Benchmark: vklBenchmark ISPC
OpenVKL
Benchmark: vklBenchmark Scalar
LuxCoreRender
Scene: LuxCore Benchmark - Acceleration: CPU
simdjson
Throughput Test: TopTweet
simdjson
Throughput Test: LargeRandom
simdjson
Throughput Test: PartialTweets
PyPerformance
Benchmark: pickle_pure_python
PyPerformance
Benchmark: json_loads
PyPerformance
Benchmark: crypto_pyaes
PyPerformance
Benchmark: chaos
Selenium
Benchmark: WASM imageConvolute - Browser: Firefox
Selenium
Benchmark: Speedometer - Browser: Google Chrome
NCNN
Target: CPU - Model: FastestDet
NCNN
Target: CPU - Model: regnety_400m
NCNN
Target: CPU - Model: squeezenet_ssd
NCNN
Target: CPU - Model: blazeface
NCNN
Target: CPU - Model: efficientnet-b0
NCNN
Target: CPU - Model: mnasnet
NCNN
Target: CPU - Model: shufflenet-v2
NCNN
Target: CPU-v3-v3 - Model: mobilenet-v3
NCNN
Target: CPU-v2-v2 - Model: mobilenet-v2
NCNN
Target: CPU - Model: mobilenet
Cryptsetup
AES-XTS 256b Encryption
Pennant
Test: sedovbig
FinanceBench
Benchmark: Bonds OpenMP
Redis
Test: GET - Parallel Connections: 50
Redis
Test: SET - Parallel Connections: 500
Phoronix Test Suite v10.8.5