HPC - High Performance Computing HPC - High Performance Computing

A collection of common HPC (High Performance Computing) benchmarks.

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

Tests In This Suite

  • ACES DGEMM

  • AI Benchmark Alpha

  • Algebraic Multi-Grid Benchmark

  • ArrayFire

  •         Test: BLAS CPU
  • ASKAP

  •         Test: tConvolve CUDA
  •         Test: tConvolve MPI
  •         Test: tConvolve OpenMP
  •         Test: tConvolve MT
  •         Test: Hogbom Clean OpenMP
  •         Test: tConvolve OpenCL
  • Caffe

  •         Model: AlexNet - Acceleration: CPU - Iterations: 1000
  •         Model: AlexNet - Acceleration: CPU - Iterations: 100
  •         Model: AlexNet - Acceleration: CPU - Iterations: 200
  •         Model: GoogleNet - Acceleration: CPU - Iterations: 200
  •         Model: GoogleNet - Acceleration: CPU - Iterations: 100
  •         Model: GoogleNet - Acceleration: CPU - Iterations: 1000
  • CloverLeaf

  • CP2K Molecular Dynamics

  • Darmstadt Automotive Parallel Heterogeneous Suite

  •         Backend: NVIDIA CUDA - Kernel: Points2Image
  •         Backend: OpenMP - Kernel: NDT Mapping
  •         Backend: OpenMP - Kernel: Points2Image
  •         Backend: OpenCL - Kernel: Euclidean Cluster
  •         Backend: OpenCL - Kernel: NDT Mapping
  •         Backend: OpenCL - Kernel: Points2Image
  •         Backend: OpenMP - Kernel: Euclidean Cluster
  •         Backend: NVIDIA CUDA - Kernel: Euclidean Cluster
  •         Backend: NVIDIA CUDA - Kernel: NDT Mapping
  • DeepSpeech

  •         Acceleration: CPU
  • Dolfyn

  • ECP-CANDLE

  •         Benchmark: P3B2
  •         Benchmark: P3B1
  •         Benchmark: P1B2
  • FFTE

  •         Test: N=256, 1D Complex FFT Routine
  • FFTW

  •         Build: Stock - Size: 2D FFT Size 4096
  •         Build: Float + SSE - Size: 2D FFT Size 32
  •         Build: Float + SSE - Size: 1D FFT Size 4096
  •         Build: Float + SSE - Size: 1D FFT Size 32
  •         Build: Stock - Size: 1D FFT Size 32
  •         Build: Stock - Size: 1D FFT Size 4096
  •         Build: Stock - Size: 2D FFT Size 32
  •         Build: Float + SSE - Size: 2D FFT Size 4096
  • GNU Octave Benchmark

  • GPAW

  •         Input: Carbon Nanotube
  • GROMACS

  •         Input: water_GMX50_bare
  • High Performance Conjugate Gradient

  • Himeno Benchmark

  • HPC Challenge

  •         Test / Class: G-HPL
  •         Test / Class: EP-DGEMM
  •         Test / Class: G-Ptrans
  •         Test / Class: G-Random Access
  •         Test / Class: G-Ffte
  •         Test / Class: EP-STREAM Triad
  •         Test / Class: Max Ping Pong Bandwidth
  •         Test / Class: Random Ring Bandwidth
  •         Test / Class: Random Ring Latency
  • Incompact3D

  •         Input: Cylinder
  • Intel MPI Benchmarks

  •         Test: IMB-MPI1 PingPong
  •         Test: IMB-MPI1 Sendrecv
  •         Test: IMB-MPI1 Exchange
  •         Test: IMB-P2P PingPong
  • Kripke

  • LAMMPS Molecular Dynamics Simulator

  •         Model: 20k Atoms
  •         Model: Rhodopsin Protein
  • LeelaChessZero

  •         Backend: BLAS
  • LULESH

  • miniFE

  •         Problem Size: Small
  •         Problem Size: Medium
  •         Problem Size: Large
  • Mlpack Benchmark

  •         Benchmark: scikit_svm
  •         Benchmark: scikit_qda
  •         Benchmark: scikit_ica
  •         Benchmark: scikit_linearridgeregression
  • Mobile Neural Network

  • Monte Carlo Simulations of Ionised Nebulae

  • NAMD

  • NAS Parallel Benchmarks

  •         Test / Class: EP.C
  •         Test / Class: BT.C
  •         Test / Class: LU.C
  •         Test / Class: FT.C
  •         Test / Class: SP.B
  •         Test / Class: EP.D
  •         Test / Class: IS.D
  •         Test / Class: MG.C
  •         Test / Class: CG.C
  • NCNN

  •         Target: CPU
  •         Target: Vulkan GPU
  • Nebular Empirical Analysis Tool

  • Numenta Anomaly Benchmark

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

  • NWChem

  • oneDNN

  •         Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU
  •         Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU
  •         Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU
  •         Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU
  •         Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU
  •         Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU
  •         Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
  •         Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU
  •         Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU
  •         Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU
  •         Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU
  •         Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU
  •         Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU
  •         Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU
  •         Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU
  • ONNX Runtime

  •         Model: yolov4 - Device: OpenMP CPU
  •         Model: fcn-resnet101-11 - Device: OpenMP CPU
  •         Model: shufflenet-v2-10 - Device: OpenMP CPU
  •         Model: super-resolution-10 - Device: OpenMP CPU
  •         Model: bertsquad-10 - Device: OpenMP CPU
  • OpenCV

  •         Test: DNN - Deep Neural Network
  • OpenFOAM

  •         Input: Motorbike 60M
  •         Input: Motorbike 30M
  • OpenVINO

  •         Model: Age Gender Recognition Retail 0013 FP16 - Device: Intel GPU
  •         Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU
  •         Model: Face Detection 0106 FP16 - Device: CPU
  •         Model: Face Detection 0106 FP16 - Device: Intel GPU
  •         Model: Face Detection 0106 FP32 - Device: CPU
  •         Model: Face Detection 0106 FP32 - Device: Intel GPU
  •         Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
  •         Model: Person Detection 0106 FP32 - Device: Intel GPU
  •         Model: Person Detection 0106 FP32 - Device: CPU
  •         Model: Person Detection 0106 FP16 - Device: Intel GPU
  •         Model: Person Detection 0106 FP16 - Device: CPU
  •         Model: Age Gender Recognition Retail 0013 FP32 - Device: Intel GPU
  • Parboil

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

  •         Test: sedovbig
  •         Test: leblancbig
  • PlaidML

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

  •         Input: simple-H2O
  • Quantum ESPRESSO

  •         Input: AUSURF112
  • R Benchmark

  • RELION

  •         Test: Basic - Device: CPU
  • RNNoise

  • Rodinia

  •         Test: OpenMP LavaMD
  •         Test: OpenMP CFD Solver
  •         Test: OpenMP Leukocyte
  •         Test: OpenMP Streamcluster
  • Scikit-Learn

  • SHOC Scalable HeterOgeneous Computing

  •         Target: CUDA - Benchmark: Bus Speed Download
  •         Target: OpenCL - Benchmark: Bus Speed Download
  •         Target: OpenCL - Benchmark: Triad
  •         Target: OpenCL - Benchmark: FFT SP
  •         Target: OpenCL - Benchmark: Texture Read Bandwidth
  •         Target: OpenCL - Benchmark: Max SP Flops
  •         Target: OpenCL - Benchmark: Bus Speed Readback
  •         Target: OpenCL - Benchmark: MD5 Hash
  •         Target: CUDA - Benchmark: Triad
  •         Target: CUDA - Benchmark: MD5 Hash
  •         Target: CUDA - Benchmark: FFT SP
  •         Target: CUDA - Benchmark: Texture Read Bandwidth
  •         Target: CUDA - Benchmark: Max SP Flops
  •         Target: CUDA - Benchmark: Bus Speed Readback
  • Tensorflow

  •         Build: Cifar10
  • TensorFlow Lite

  •         Model: Mobilenet Float
  •         Model: Mobilenet Quant
  •         Model: NASNet Mobile
  •         Model: SqueezeNet
  •         Model: Inception ResNet V2
  •         Model: Inception V4
  • Timed HMMer Search

  • Timed MAFFT Alignment

  • Timed MrBayes Analysis

  • TNN

  •         Target: CPU - Model: SqueezeNet v1.1
  •         Target: CPU - Model: MobileNet v2

Revision History Revision History

pts/hpc-1.1.1     Mon, 18 Jan 2021 21:03:32 GMT
Add IOR to HPC suite for disk / I/O performance.

pts/hpc-1.1.0     Thu, 14 Jan 2021 13:53:57 GMT
Add new tests.

pts/hpc-1.0.0     Wed, 08 Apr 2020 15:08:51 GMT
Initial commit of HPC high performance computing benchmarks suite for easy access.