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
- Test: BLAS CPU
- Test: tConvolve OpenCL
- Test: tConvolve CUDA
- Test: tConvolve MPI
- Test: tConvolve OpenMP
- Test: tConvolve MT
- Test: Hogbom Clean OpenMP
- Model: AlexNet - Acceleration: CPU - Iterations: 100
- Model: AlexNet - Acceleration: CPU - Iterations: 200
- Model: AlexNet - Acceleration: CPU - Iterations: 1000
- Model: GoogleNet - Acceleration: CPU - Iterations: 100
- Model: GoogleNet - Acceleration: CPU - Iterations: 200
- Model: GoogleNet - Acceleration: CPU - Iterations: 1000
- Input: clover_bm
- Input: clover_bm64_short
- Input: clover_bm16
- Input: Fayalite-FIST
- Input: H20-64
- Input: H20-256
- Backend: OpenMP - Kernel: Euclidean Cluster
- Backend: OpenMP - Kernel: NDT Mapping
- Backend: OpenMP - Kernel: Points2Image
- Backend: OpenCL - Kernel: Euclidean Cluster
- Backend: OpenCL - Kernel: NDT Mapping
- Backend: OpenCL - Kernel: Points2Image
- Backend: NVIDIA CUDA - Kernel: Euclidean Cluster
- Backend: NVIDIA CUDA - Kernel: NDT Mapping
- Backend: NVIDIA CUDA - Kernel: Points2Image
- Acceleration: CPU
- Input: e2Asean Grid + BengkuluSept2007 Source - Time: 240
- Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200
- Input: e2Asean Grid + BengkuluSept2007 Source - Time: 2400
- Benchmark: P1B2
- Benchmark: P3B1
- Benchmark: P3B2
- Epoch3D Deck: Cone
- Test: demo_sift1M
- Test: bench_polysemous_sift1m
- Test: N=256, 1D Complex FFT Routine
- Build: Stock - Size: 1D FFT Size 32
- Build: Stock - Size: 1D FFT Size 4096
- Build: Stock - Size: 2D FFT Size 32
- Build: Stock - Size: 2D FFT Size 4096
- Build: Float + SSE - Size: 1D FFT Size 32
- Build: Float + SSE - Size: 1D FFT Size 4096
- Build: Float + SSE - Size: 2D FFT Size 32
- Build: Float + SSE - Size: 2D FFT Size 4096
- Input: Carbon Nanotube
- Scale: 26
- Scale: 29
- Implementation: MPI CPU - Input: water_GMX50_bare
- Implementation: NVIDIA CUDA GPU - Input: water_GMX50_bare
- Test: r2c - Backend: Stock - Precision: float - X Y Z: 128
- Test: r2c - Backend: Stock - Precision: float - X Y Z: 256
- Test: r2c - Backend: Stock - Precision: float - X Y Z: 512
- Test: r2c - Backend: Stock - Precision: float - X Y Z: 1024
- Test: r2c - Backend: Stock - Precision: float-long - X Y Z: 128
- Test: r2c - Backend: Stock - Precision: float-long - X Y Z: 256
- Test: r2c - Backend: Stock - Precision: float-long - X Y Z: 512
- Test: r2c - Backend: Stock - Precision: float-long - X Y Z: 1024
- Test: r2c - Backend: Stock - Precision: double - X Y Z: 128
- Test: r2c - Backend: Stock - Precision: double - X Y Z: 256
- Test: r2c - Backend: Stock - Precision: double - X Y Z: 512
- Test: r2c - Backend: Stock - Precision: double - X Y Z: 1024
- Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 128
- Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 256
- Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 512
- Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 1024
- Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128
- Test: r2c - Backend: FFTW - Precision: float - X Y Z: 256
- Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512
- Test: r2c - Backend: FFTW - Precision: float - X Y Z: 1024
- Test: r2c - Backend: FFTW - Precision: float-long - X Y Z: 128
- Test: r2c - Backend: FFTW - Precision: float-long - X Y Z: 256
- Test: r2c - Backend: FFTW - Precision: float-long - X Y Z: 512
- Test: r2c - Backend: FFTW - Precision: float-long - X Y Z: 1024
- Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128
- Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256
- Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512
- Test: r2c - Backend: FFTW - Precision: double - X Y Z: 1024
- Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 128
- Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 256
- Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 512
- Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 1024
- Test: c2c - Backend: Stock - Precision: float - X Y Z: 128
- Test: c2c - Backend: Stock - Precision: float - X Y Z: 256
- Test: c2c - Backend: Stock - Precision: float - X Y Z: 512
- Test: c2c - Backend: Stock - Precision: float - X Y Z: 1024
- Test: c2c - Backend: Stock - Precision: float-long - X Y Z: 128
- Test: c2c - Backend: Stock - Precision: float-long - X Y Z: 256
- Test: c2c - Backend: Stock - Precision: float-long - X Y Z: 512
- Test: c2c - Backend: Stock - Precision: float-long - X Y Z: 1024
- Test: c2c - Backend: Stock - Precision: double - X Y Z: 128
- Test: c2c - Backend: Stock - Precision: double - X Y Z: 256
- Test: c2c - Backend: Stock - Precision: double - X Y Z: 512
- Test: c2c - Backend: Stock - Precision: double - X Y Z: 1024
- Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 128
- Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 256
- Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 512
- Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 1024
- Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128
- Test: c2c - Backend: FFTW - Precision: float - X Y Z: 256
- Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512
- Test: c2c - Backend: FFTW - Precision: float - X Y Z: 1024
- Test: c2c - Backend: FFTW - Precision: float-long - X Y Z: 128
- Test: c2c - Backend: FFTW - Precision: float-long - X Y Z: 256
- Test: c2c - Backend: FFTW - Precision: float-long - X Y Z: 512
- Test: c2c - Backend: FFTW - Precision: float-long - X Y Z: 1024
- Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128
- Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256
- Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512
- Test: c2c - Backend: FFTW - Precision: double - X Y Z: 1024
- Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 128
- Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 256
- Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 512
- Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 1024
- X Y Z: 104 104 104 - RT: 60
- X Y Z: 104 104 104 - RT: 1800
- X Y Z: 144 144 144 - RT: 60
- X Y Z: 144 144 144 - RT: 1800
- X Y Z: 160 160 160 - RT: 60
- X Y Z: 160 160 160 - RT: 1800
- X Y Z: 192 192 192 - RT: 60
- X Y Z: 192 192 192 - RT: 1800
- Test / Class: G-HPL
- Test / Class: G-Ptrans
- Test / Class: G-Random Access
- Test / Class: G-Ffte
- Test / Class: EP-STREAM Triad
- Test / Class: EP-DGEMM
- Test / Class: Random Ring Latency
- Test / Class: Random Ring Bandwidth
- Test / Class: Max Ping Pong Bandwidth
- Test: IMB-MPI1 PingPong
- Test: IMB-MPI1 Sendrecv
- Test: IMB-MPI1 Exchange
- Test: IMB-P2P PingPong
- Test: Sedov Blast Wave, ube_922_hex.mesh
- Test: Triple Point Problem
- Model: Rhodopsin Protein
- Model: 20k Atoms
- Backend: BLAS
- M N K: 32
- M N K: 64
- M N K: 128
- M N K: 256
- Model: Mobilenet Float
- Model: Mobilenet Quant
- Model: NASNet Mobile
- Model: SqueezeNet
- Model: Inception ResNet V2
- Model: Inception V4
- Model: Quantized COCO SSD MobileNet v1
- Model: DeepLab V3
- Model: llama-2-7b.Q4_0.gguf
- Model: llama-2-13b.Q4_0.gguf
- Model: llama-2-70b-chat.Q5_0.gguf
- Test: mistral-7b-instruct-v0.2.Q8_0 - Acceleration: CPU
- Test: llava-v1.5-7b-q4 - Acceleration: CPU
- Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU
- Implementation: OpenMP - Input Deck: BM1
- Implementation: OpenMP - Input Deck: BM2
- Problem Size: Small
- Problem Size: Medium
- Problem Size: Large
- Benchmark: scikit_svm
- Benchmark: scikit_linearridgeregression
- Benchmark: scikit_qda
- Benchmark: scikit_ica
- Input: Dust 2D tau100.0
- Input: Gas HII40
- Input: ATPase with 327,506 Atoms
- Input: STMV with 1,066,628 Atoms
- Test / Class: BT.C
- Test / Class: EP.C
- Test / Class: EP.D
- Test / Class: FT.C
- Test / Class: LU.C
- Test / Class: SP.B
- Test / Class: SP.C
- Test / Class: IS.D
- Test / Class: MG.C
- Test / Class: CG.C
- Target: CPU
- Target: Vulkan GPU
- Input: TurboPipe Periodic
- Input: Kershaw
- Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream
- Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream
- Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream
- Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
- Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream
- Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
- Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream
- Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
- Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream
- Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream
- Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream
- Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
- Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream
- Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
- Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream
- Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream
- Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream
- Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream
- Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream
- Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream
- Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Stream
- Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Stream
- Detector: Bayesian Changepoint
- Detector: Windowed Gaussian
- Detector: Relative Entropy
- Detector: Earthgecko Skyline
- Detector: KNN CAD
- Detector: Contextual Anomaly Detector OSE
- Input: C240 Buckyball
- Harness: Convolution Batch Shapes Auto - Engine: CPU
- Harness: Deconvolution Batch shapes_1d - Engine: CPU
- Harness: Deconvolution Batch shapes_3d - Engine: CPU
- Harness: IP Shapes 1D - Engine: CPU
- Harness: IP Shapes 3D - Engine: CPU
- Harness: Recurrent Neural Network Training - Engine: CPU
- Harness: Recurrent Neural Network Inference - Engine: CPU
- Model: yolov4 - Device: CPU - Executor: Standard
- Model: yolov4 - Device: CPU - Executor: Parallel
- Model: fcn-resnet101-11 - Device: CPU - Executor: Standard
- Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel
- Model: super-resolution-10 - Device: CPU - Executor: Standard
- Model: super-resolution-10 - Device: CPU - Executor: Parallel
- Model: bertsquad-12 - Device: CPU - Executor: Standard
- Model: bertsquad-12 - Device: CPU - Executor: Parallel
- Model: GPT-2 - Device: CPU - Executor: Standard
- Model: GPT-2 - Device: CPU - Executor: Parallel
- Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard
- Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel
- Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard
- Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel
- Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard
- Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel
- Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard
- Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel
- Model: T5 Encoder - Device: CPU - Executor: Standard
- Model: T5 Encoder - Device: CPU - Executor: Parallel
- Model: ZFNet-512 - Device: CPU - Executor: Standard
- Model: ZFNet-512 - Device: CPU - Executor: Parallel
- Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard
- Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel
- Test: DNN - Deep Neural Network
- Input: motorBike
- Input: drivaerFastback, Small Mesh Size
- Input: drivaerFastback, Medium Mesh Size
- Input: drivaerFastback, Large Mesh Size
- Model: Bird Strike on Windshield
- Model: Rubber O-Ring Seal Installation
- Model: Cell Phone Drop Test
- Model: Bumper Beam
- Model: INIVOL and Fluid Structure Interaction Drop Container
- Model: Chrysler Neon 1M
- Model: Ford Taurus 10M
- Model: Face Detection FP16 - Device: CPU
- Model: Face Detection FP16-INT8 - Device: CPU
- Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
- Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU
- Model: Person Detection FP16 - Device: CPU
- Model: Person Detection FP32 - Device: CPU
- Model: Weld Porosity Detection FP16-INT8 - Device: CPU
- Model: Weld Porosity Detection FP16 - Device: CPU
- Model: Vehicle Detection FP16-INT8 - Device: CPU
- Model: Vehicle Detection FP16 - Device: CPU
- Model: Person Vehicle Bike Detection FP16 - Device: CPU
- Model: Machine Translation EN To DE FP16 - Device: CPU
- Model: Face Detection Retail FP16 - Device: CPU
- Model: Face Detection Retail FP16-INT8 - Device: CPU
- Model: Handwritten English Recognition FP16 - Device: CPU
- Model: Handwritten English Recognition FP16-INT8 - Device: CPU
- Model: Road Segmentation ADAS FP16 - Device: CPU
- Model: Road Segmentation ADAS FP16-INT8 - Device: CPU
- Model: Person Re-Identification Retail FP16 - Device: CPU
- Model: Noise Suppression Poconet-Like FP16 - Device: CPU
- Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU
- Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU
- Model: Falcon-7b-instruct-int4-ov - Device: CPU
- Model: Gemma-7b-int4-ov - Device: CPU
- Grid Size: 100
- Grid Size: 400
- Grid Size: 500
- Grid Size: 1000
- Grid Size: 4000
- Test: OpenMP CUTCP
- Test: OpenMP MRI-Q
- Test: OpenMP MRI Gridding
- Test: OpenMP Stencil
- Test: OpenMP LBM
- Test: leblancbig
- Test: sedovbig
- Test: Streams
- FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
- FP16: No - Mode: Inference - Network: VGG16 - Device: CPU
- Device: CPU - Backend: Aesara - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: Aesara - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Aesara - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: Aesara - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Aesara - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: Aesara - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Aesara - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: Aesara - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Aesara - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: Aesara - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: JAX - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: JAX - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: JAX - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: JAX - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: JAX - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: JAX - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: JAX - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: JAX - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: JAX - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: JAX - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numba - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: Numba - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numba - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: Numba - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numba - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: Numba - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numba - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: Numba - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: Numba - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: Numba - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: PyTorch - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: PyTorch - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: PyTorch - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: PyTorch - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: PyTorch - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: PyTorch - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: PyTorch - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: PyTorch - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: PyTorch - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: PyTorch - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: TensorFlow - Project Size: 16384 - Benchmark: Equation of State
- Device: CPU - Backend: TensorFlow - Project Size: 16384 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: TensorFlow - Project Size: 65536 - Benchmark: Equation of State
- Device: CPU - Backend: TensorFlow - Project Size: 65536 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: TensorFlow - Project Size: 262144 - Benchmark: Equation of State
- Device: CPU - Backend: TensorFlow - Project Size: 262144 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: TensorFlow - Project Size: 1048576 - Benchmark: Equation of State
- Device: CPU - Backend: TensorFlow - Project Size: 1048576 - Benchmark: Isoneutral Mixing
- Device: CPU - Backend: TensorFlow - Project Size: 4194304 - Benchmark: Equation of State
- Device: CPU - Backend: TensorFlow - Project Size: 4194304 - Benchmark: Isoneutral Mixing
- Device: CPU - Batch Size: 1 - Model: ResNet-50
- Device: CPU - Batch Size: 1 - Model: ResNet-152
- Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l
- Device: CPU - Batch Size: 16 - Model: ResNet-50
- Device: CPU - Batch Size: 16 - Model: ResNet-152
- Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l
- Device: CPU - Batch Size: 32 - Model: ResNet-50
- Device: CPU - Batch Size: 32 - Model: ResNet-152
- Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l
- Device: CPU - Batch Size: 64 - Model: ResNet-50
- Device: CPU - Batch Size: 64 - Model: ResNet-152
- Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l
- Device: CPU - Batch Size: 256 - Model: ResNet-50
- Device: CPU - Batch Size: 256 - Model: ResNet-152
- Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l
- Device: CPU - Batch Size: 512 - Model: ResNet-50
- Device: CPU - Batch Size: 512 - Model: ResNet-152
- Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l
- Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50
- Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152
- Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l
- Input: simple-H2O
- Input: Li2_STO_ae
- Input: FeCO6_b3lyp_gms
- Input: O_ae_pyscf_UHF
- Input: LiH_ae_MSD
- Input: H4_ae
- Input: AUSURF112
- Input: CORAL2 P1
- Input: CORAL2 P2
- Input: CTS2
- Test: Basic - Device: CPU
- Test: Sample Remap Example
- Input: 26 Minute Long Talking Sample
- Test: OpenMP CFD Solver
- Test: OpenMP LavaMD
- Test: OpenMP Leukocyte
- Test: OpenMP Streamcluster
- Benchmark: 20 Newsgroups / Logistic Regression
- Benchmark: Covertype Dataset Benchmark
- Benchmark: Feature Expansions
- Benchmark: GLM
- Benchmark: Glmnet
- Benchmark: Hist Gradient Boosting
- Benchmark: Hist Gradient Boosting Adult
- Benchmark: Hist Gradient Boosting Categorical Only
- Benchmark: Hist Gradient Boosting Higgs Boson
- Benchmark: Hist Gradient Boosting Threading
- Benchmark: Isolation Forest
- Benchmark: Isotonic / Perturbed Logarithm
- Benchmark: Isotonic / Logistic
- Benchmark: Isotonic / Pathological
- Benchmark: Kernel PCA Solvers / Time vs. N Components
- Benchmark: Kernel PCA Solvers / Time vs. N Samples
- Benchmark: Lasso
- Benchmark: LocalOutlierFactor
- Benchmark: SGDOneClassSVM
- Benchmark: Plot Fast KMeans
- Benchmark: Plot Hierarchical
- Benchmark: Plot Incremental PCA
- Benchmark: Plot Lasso Path
- Benchmark: Plot Neighbors
- Benchmark: Plot Non-Negative Matrix Factorization
- Benchmark: Plot OMP vs. LARS
- Benchmark: Plot Parallel Pairwise
- Benchmark: Plot Polynomial Kernel Approximation
- Benchmark: Plot Singular Value Decomposition
- Benchmark: Plot Ward
- Benchmark: Sparse Random Projections / 100 Iterations
- Benchmark: RCV1 Logreg Convergencet
- Benchmark: SAGA
- Benchmark: Sample Without Replacement
- Benchmark: SGD Regression
- Benchmark: Sparsify
- Benchmark: Text Vectorizers
- Benchmark: Tree
- Benchmark: MNIST Dataset
- Benchmark: TSNE MNIST Dataset
- Target: OpenCL - Benchmark: Bus Speed Download
- Target: OpenCL - Benchmark: Bus Speed Readback
- Target: OpenCL - Benchmark: Max SP Flops
- Target: OpenCL - Benchmark: Texture Read Bandwidth
- Target: OpenCL - Benchmark: FFT SP
- Target: OpenCL - Benchmark: GEMM SGEMM_N
- Target: OpenCL - Benchmark: MD5 Hash
- Target: OpenCL - Benchmark: Reduction
- Target: OpenCL - Benchmark: Triad
- Target: OpenCL - Benchmark: S3D
- Model: Layered Halfspace
- Model: Water-layered Halfspace
- Model: Homogeneous Halfspace
- Model: Mount St. Helens
- Model: Tomographic Model
- Device: CPU - Batch Size: 1 - Model: VGG-16
- Device: CPU - Batch Size: 1 - Model: ResNet-50
- Device: CPU - Batch Size: 1 - Model: AlexNet
- Device: CPU - Batch Size: 1 - Model: GoogLeNet
- Device: CPU - Batch Size: 16 - Model: VGG-16
- Device: CPU - Batch Size: 16 - Model: ResNet-50
- Device: CPU - Batch Size: 16 - Model: AlexNet
- Device: CPU - Batch Size: 16 - Model: GoogLeNet
- Device: CPU - Batch Size: 32 - Model: VGG-16
- Device: CPU - Batch Size: 32 - Model: ResNet-50
- Device: CPU - Batch Size: 32 - Model: AlexNet
- Device: CPU - Batch Size: 32 - Model: GoogLeNet
- Device: CPU - Batch Size: 64 - Model: VGG-16
- Device: CPU - Batch Size: 64 - Model: ResNet-50
- Device: CPU - Batch Size: 64 - Model: AlexNet
- Device: CPU - Batch Size: 64 - Model: GoogLeNet
- Device: CPU - Batch Size: 256 - Model: VGG-16
- Device: CPU - Batch Size: 256 - Model: ResNet-50
- Device: CPU - Batch Size: 256 - Model: AlexNet
- Device: CPU - Batch Size: 256 - Model: GoogLeNet
- Device: CPU - Batch Size: 512 - Model: VGG-16
- Device: CPU - Batch Size: 512 - Model: ResNet-50
- Device: CPU - Batch Size: 512 - Model: AlexNet
- Device: CPU - Batch Size: 512 - Model: GoogLeNet
- Device: GPU - Batch Size: 1 - Model: VGG-16
- Device: GPU - Batch Size: 1 - Model: ResNet-50
- Device: GPU - Batch Size: 1 - Model: AlexNet
- Device: GPU - Batch Size: 1 - Model: GoogLeNet
- Device: GPU - Batch Size: 16 - Model: VGG-16
- Device: GPU - Batch Size: 16 - Model: ResNet-50
- Device: GPU - Batch Size: 16 - Model: AlexNet
- Device: GPU - Batch Size: 16 - Model: GoogLeNet
- Device: GPU - Batch Size: 32 - Model: VGG-16
- Device: GPU - Batch Size: 32 - Model: ResNet-50
- Device: GPU - Batch Size: 32 - Model: AlexNet
- Device: GPU - Batch Size: 32 - Model: GoogLeNet
- Device: GPU - Batch Size: 64 - Model: VGG-16
- Device: GPU - Batch Size: 64 - Model: ResNet-50
- Device: GPU - Batch Size: 64 - Model: AlexNet
- Device: GPU - Batch Size: 64 - Model: GoogLeNet
- Device: GPU - Batch Size: 256 - Model: VGG-16
- Device: GPU - Batch Size: 256 - Model: ResNet-50
- Device: GPU - Batch Size: 256 - Model: AlexNet
- Device: GPU - Batch Size: 256 - Model: GoogLeNet
- Device: GPU - Batch Size: 512 - Model: VGG-16
- Device: GPU - Batch Size: 512 - Model: ResNet-50
- Device: GPU - Batch Size: 512 - Model: AlexNet
- Device: GPU - Batch Size: 512 - Model: GoogLeNet
- Model: Mobilenet Float
- Model: Mobilenet Quant
- Model: NASNet Mobile
- Model: SqueezeNet
- Model: Inception ResNet V2
- Model: Inception V4
- Target: CPU - Model: DenseNet
- Target: CPU - Model: MobileNet v2
- Target: CPU - Model: SqueezeNet v1.1
- Target: CPU - Model: SqueezeNet v2
- Input: Plasma Acceleration
- Input: Uniform Plasma
- Model: ggml-base.en - Input: 2016 State of the Union
- Model: ggml-small.en - Input: 2016 State of the Union
- Model: ggml-medium.en - Input: 2016 State of the Union
- Model Size: Tiny
- Model Size: Small
- Model Size: Medium
- Input: conus 2.5km
- Input: input.i3d 129 Cells Per Direction
- Input: input.i3d 193 Cells Per Direction
- Input: X3D-benchmarking input.i3d
Revision History
pts/hpc-1.1.12 Sun, 24 Nov 2024 10:47:23 GMT
Add warpx and epoch to HPC test suite.
pts/hpc-1.1.11 Sat, 27 Jan 2024 19:48:56 GMT
Add QuickSilver to HPC suite.
pts/hpc-1.1.10 Sat, 14 Oct 2023 20:33:55 GMT
Add easywave to test suite
pts/hpc-1.1.9 Sun, 06 Aug 2023 17:01:43 GMT
Add libxsmm, heffte, remhos, laghos, palabos, faiss, and petsc tests to HPC suite.
pts/hpc-1.1.8 Sat, 25 Mar 2023 18:15:31 GMT
Add specfem3d to HPC suite.
pts/hpc-1.1.7 Fri, 18 Nov 2022 14:55:03 GMT
Add nekrs and minibude to HPC test suite
pts/hpc-1.1.6 Thu, 13 Oct 2022 16:47:18 GMT
Add OpenRadioss to HPC test suite.
pts/hpc-1.1.5 Fri, 28 Jan 2022 07:50:31 GMT
Add Graph500 to HPC suite.
pts/hpc-1.1.4 Fri, 22 Oct 2021 15:23:06 GMT
Add PyHPC to HPC test suite.
pts/hpc-1.1.3 Tue, 27 Apr 2021 18:01:43 GMT
Add WRF to HPC test suite.
pts/hpc-1.1.2 Tue, 09 Mar 2021 11:24:37 GMT
Add hpl to suite.
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.