Clear Linux vs. Ubuntu 21.04 Benchmarks

Ubuntu 21.04 vs. Clear Linux benchmarks for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2104228-IB-2104227IB95&grr.

PlaidML

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

LeelaChessZero

Backend: Eigen

Blender

Blend File: Barbershop - Compute: CPU-Only

PlaidML

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

AI Benchmark Alpha

Device AI Score

AI Benchmark Alpha

Device Training Score

AI Benchmark Alpha

Device Inference Score

PlaidML

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

SecureMark

Benchmark: SecureMark-TLS

Selenium

Benchmark: Jetstream 2 - Browser: Firefox

Selenium

Benchmark: StyleBench - Browser: Firefox

NAMD

ATPase Simulation - 327,506 Atoms

PlaidML

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

NAS Parallel Benchmarks

Test / Class: EP.D

GNU GMP GMPbench

Total Time

Rodinia

Test: OpenMP LavaMD

GROMACS

Input: water_GMX50_bare

OSPray

Demo: San Miguel - Renderer: Path Tracer

TensorFlow Lite

Model: Inception V4

Selenium

Benchmark: Speedometer - Browser: Firefox

PlaidML

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

OpenVKL

Benchmark: vklBenchmark

Blender

Blend File: BMW27 - Compute: CPU-Only

Rodinia

Test: OpenMP HotSpot3D

OSPray

Demo: XFrog Forest - Renderer: Path Tracer

PlaidML

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

Selenium

Benchmark: WASM imageConvolute - Browser: Firefox

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

Rodinia

Test: OpenMP Streamcluster

TensorFlow Lite

Model: Inception ResNet V2

PyPerformance

Benchmark: python_startup

Chaos Group V-RAY

Mode: CPU

simdjson

Throughput Test: PartialTweets

simdjson

Throughput Test: DistinctUserID

PyPerformance

Benchmark: raytrace

IndigoBench

Acceleration: CPU - Scene: Bedroom

simdjson

Throughput Test: Kostya

IndigoBench

Acceleration: CPU - Scene: Supercar

LuxCoreRender

Scene: Orange Juice - Acceleration: CPU

LuxCoreRender

Scene: Danish Mood - Acceleration: CPU

LuxCoreRender

Scene: LuxCore Benchmark - Acceleration: CPU

Selenium

Benchmark: PSPDFKit WASM - Browser: Firefox

LuxCoreRender

Scene: DLSC - Acceleration: CPU

TensorFlow Lite

Model: SqueezeNet

TensorFlow Lite

Model: Mobilenet Quant

TensorFlow Lite

Model: NASNet Mobile

TensorFlow Lite

Model: Mobilenet Float

Rodinia

Test: OpenMP Leukocyte

simdjson

Throughput Test: LargeRandom

PyPerformance

Benchmark: 2to3

RawTherapee

Total Benchmark Time

Timed Linux Kernel Compilation

Time To Compile

NAS Parallel Benchmarks

Test / Class: LU.C

OSPray

Demo: XFrog Forest - Renderer: SciVis

QuantLib

OpenVKL

Benchmark: vklBenchmarkStructuredVolume

SVT-HEVC

Tuning: 1 - Input: Bosphorus 1080p

Selenium

Benchmark: Kraken - Browser: Firefox

PyPerformance

Benchmark: go

LuxCoreRender

Scene: Rainbow Colors and Prism - Acceleration: CPU

PyPerformance

Benchmark: float

OSPray

Demo: NASA Streamlines - Renderer: Path Tracer

PyPerformance

Benchmark: django_template

Embree

Binary: Pathtracer - Model: Crown

PyPerformance

Benchmark: chaos

PyPerformance

Benchmark: crypto_pyaes

Stockfish

Total Time

LuaJIT

Test: Composite

OSPray

Demo: San Miguel - Renderer: SciVis

Embree

Binary: Pathtracer ISPC - Model: Crown

PyPerformance

Benchmark: regex_compile

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU

x265

Video Input: Bosphorus 4K

Embree

Binary: Pathtracer - Model: Asian Dragon

Selenium

Benchmark: WASM collisionDetection - Browser: Firefox

PyPerformance

Benchmark: nbody

PyPerformance

Benchmark: json_loads

PyPerformance

Benchmark: pathlib

Rodinia

Test: OpenMP CFD Solver

PHPBench

PHP Benchmark Suite

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

PyPerformance

Benchmark: pickle_pure_python

PyBench

Total For Average Test Times

Liquid-DSP

Threads: 36 - Buffer Length: 256 - Filter Length: 57

Liquid-DSP

Threads: 16 - Buffer Length: 256 - Filter Length: 57

Liquid-DSP

Threads: 32 - Buffer Length: 256 - Filter Length: 57

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

NAS Parallel Benchmarks

Test / Class: EP.C

Google Draco

Model: Church Facade

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

Intel Open Image Denoise

Scene: Memorial

OSPray

Demo: NASA Streamlines - Renderer: SciVis

OSPray

Demo: Magnetic Reconnection - Renderer: SciVis

Google Draco

Model: Lion

OSPray

Demo: Magnetic Reconnection - Renderer: Path Tracer

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

Systemd Total Boot Time

Test: Kernel

Phoronix Test Suite v10.8.5