Amazon EC2 c7g.4xlarge AWS Graviton3
Graviton3 benchmarks by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2205269-NE-2205259NE55&grs.
NAS Parallel Benchmarks
Test / Class: LU.C
ONNX Runtime
Model: fcn-resnet101-11 - Device: CPU - Executor: Standard
NAS Parallel Benchmarks
Test / Class: SP.C
NAS Parallel Benchmarks
Test / Class: MG.C
OpenSSL
Algorithm: RSA4096
Stress-NG
Test: CPU Stress
OpenSSL
Algorithm: RSA4096
NAS Parallel Benchmarks
Test / Class: FT.C
High Performance Conjugate Gradient
Timed MrBayes Analysis
Primate Phylogeny Analysis
simdjson
Throughput Test: DistinctUserID
NAS Parallel Benchmarks
Test / Class: IS.D
NAS Parallel Benchmarks
Test / Class: CG.C
ACES DGEMM
Sustained Floating-Point Rate
simdjson
Throughput Test: PartialTweets
libavif avifenc
Encoder Speed: 2
C-Ray
Total Time - 4K, 16 Rays Per Pixel
Xcompact3d Incompact3d
Input: input.i3d 193 Cells Per Direction
Stress-NG
Test: Memory Copying
ASTC Encoder
Preset: Exhaustive
ASTC Encoder
Preset: Thorough
Stress-NG
Test: Crypto
Xcompact3d Incompact3d
Input: input.i3d 129 Cells Per Direction
NAS Parallel Benchmarks
Test / Class: BT.C
simdjson
Throughput Test: Kostya
Stress-NG
Test: Matrix Math
libavif avifenc
Encoder Speed: 0
NAS Parallel Benchmarks
Test / Class: EP.D
Rodinia
Test: OpenMP LavaMD
Rodinia
Test: OpenMP CFD Solver
OpenSSL
Algorithm: SHA256
SecureMark
Benchmark: SecureMark-TLS
Algebraic Multi-Grid Benchmark
Apache HTTP Server
Concurrent Requests: 200
GROMACS
Implementation: MPI CPU - Input: water_GMX50_bare
Apache HTTP Server
Concurrent Requests: 100
PHPBench
PHP Benchmark Suite
Apache HTTP Server
Concurrent Requests: 500
LULESH
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
Ngspice
Circuit: C2670
LeelaChessZero
Backend: Eigen
simdjson
Throughput Test: LargeRandom
PyBench
Total For Average Test Times
Apache HTTP Server
Concurrent Requests: 1000
ONNX Runtime
Model: super-resolution-10 - Device: CPU - Executor: Standard
LeelaChessZero
Backend: BLAS
Zstd Compression
Compression Level: 3 - Compression Speed
7-Zip Compression
Test: Decompression Rating
Ngspice
Circuit: C7552
WebP Image Encode
Encode Settings: Quality 100, Lossless, Highest Compression
TSCP
AI Chess Performance
Timed Apache Compilation
Time To Compile
Zstd Compression
Compression Level: 19 - Decompression Speed
DaCapo Benchmark
Java Test: Tradebeans
WebP Image Encode
Encode Settings: Quality 100, Highest Compression
Zstd Compression
Compression Level: 19, Long Mode - Decompression Speed
libavif avifenc
Encoder Speed: 6, Lossless
WebP Image Encode
Encode Settings: Quality 100, Lossless
7-Zip Compression
Test: Compression Rating
Stress-NG
Test: Vector Math
Liquid-DSP
Threads: 16 - Buffer Length: 256 - Filter Length: 57
QuantLib
Timed ImageMagick Compilation
Time To Compile
libavif avifenc
Encoder Speed: 10, Lossless
Google SynthMark
Test: VoiceMark_100
DaCapo Benchmark
Java Test: Jython
Coremark
CoreMark Size 666 - Iterations Per Second
POV-Ray
Trace Time
libavif avifenc
Encoder Speed: 6
GPAW
Input: Carbon Nanotube
Timed PHP Compilation
Time To Compile
TensorFlow Lite
Model: NASNet Mobile
m-queens
Time To Solve
DaCapo Benchmark
Java Test: H2
asmFish
1024 Hash Memory, 26 Depth
Stress-NG
Test: IO_uring
TensorFlow Lite
Model: SqueezeNet
DaCapo Benchmark
Java Test: Tradesoap
Zstd Compression
Compression Level: 19, Long Mode - Compression Speed
Stockfish
Total Time
TensorFlow Lite
Model: Mobilenet Float
Timed Node.js Compilation
Time To Compile
Timed LLVM Compilation
Build System: Ninja
Timed Gem5 Compilation
Time To Compile
Build2
Time To Compile
N-Queens
Elapsed Time
Zstd Compression
Compression Level: 19 - Compression Speed
Zstd Compression
Compression Level: 3 - Decompression Speed
nginx
Concurrent Requests: 100
nginx
Concurrent Requests: 200
TensorFlow Lite
Model: Inception ResNet V2
TensorFlow Lite
Model: Inception V4
nginx
Concurrent Requests: 500
nginx
Concurrent Requests: 1000
ONNX Runtime
Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard
ONNX Runtime
Model: bertsquad-12 - Device: CPU - Executor: Standard
ONNX Runtime
Model: GPT-2 - Device: CPU - Executor: Standard
Stress-NG
Test: CPU Cache
TensorFlow Lite
Model: Mobilenet Quant
Rodinia
Test: OpenMP Streamcluster
Phoronix Test Suite v10.8.4