NVIDIA GPU Compute

A collection of test profiles that run well on NVIDIA GPU systems with CUDA / proprietary driver stack. Other deprecated / less interesting / older tests not included but this test suite is intended to serve as guidance for current interesting NVIDIA GPU compute benchmarking albeit not exhaustive of what is available via Phoronix Test Suite / OpenBenchmarking.org.

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

Tests In This Suite

  • Hashcat
        Benchmark: MD5
  • Hashcat
        Benchmark: SHA1
  • Hashcat
        Benchmark: SHA-512
  • Hashcat
        Benchmark: 7-Zip
  • Hashcat
        Benchmark: TrueCrypt RIPEMD160 + XTS
  • Waifu2x-NCNN Vulkan
        Scale: 2x - Denoise: 3 - TAA: Yes
  • Waifu2x-NCNN Vulkan
        Scale: 2x - Denoise: 3 - TAA: No
  • FAHBench
  • GROMACS
        Water Benchmark
  • NAMD CUDA
        ATPase Simulation - 327,506 Atoms
  • Mixbench
        Backend: OpenCL - Benchmark: Single Precision
  • Mixbench
        Backend: OpenCL - Benchmark: Double Precision
  • Mixbench
        Backend: OpenCL - Benchmark: Integer
  • Mixbench
        Backend: NVIDIA CUDA - Benchmark: Single Precision
  • Mixbench
        Backend: NVIDIA CUDA - Benchmark: Double Precision
  • Mixbench
        Backend: NVIDIA CUDA - Benchmark: Half Precision
  • Mixbench
        Backend: NVIDIA CUDA - Benchmark: Integer
  • OctaneBench
        Total Score
  • LuxCoreRender OpenCL
        Scene: LuxCore Benchmark
  • LuxCoreRender OpenCL
        Scene: DLSC
  • LuxCoreRender OpenCL
        Scene: Rainbow Colors and Prism
  • LuxCoreRender OpenCL
        Scene: Food
  • Rodinia
        Test: OpenCL Particle Filter
  • ArrayFire
        Test: Conjugate Gradient OpenCL
  • clpeak
        OpenCL Test: Global Memory Bandwidth
  • clpeak
        OpenCL Test: Single-Precision Float
  • clpeak
        OpenCL Test: Double-Precision Double
  • clpeak
        OpenCL Test: Integer Compute INT
  • NeatBench
        Acceleration: GPU
  • FinanceBench
        Benchmark: Black-Scholes OpenCL
  • PlaidML
        FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL
  • PlaidML
        FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL
  • PlaidML
        FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL
  • PlaidML
        FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL
  • PlaidML
        FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL
  • LeelaChessZero
        Backend: OpenCL
  • cl-mem
        Benchmark: Read
  • cl-mem
        Benchmark: Write
  • cl-mem
        Benchmark: Copy
  • MandelGPU
        OpenCL Device: GPU
  • ViennaCL
        OpenCL LU Factorization
  • RedShift Demo
  • Blender
        Blend File: BMW27 - Compute: CUDA
  • Blender
        Blend File: BMW27 - Compute: NVIDIA OptiX
  • Blender
        Blend File: Classroom - Compute: CUDA
  • Blender
        Blend File: Classroom - Compute: NVIDIA OptiX
  • Blender
        Blend File: Fishy Cat - Compute: CUDA
  • Blender
        Blend File: Fishy Cat - Compute: NVIDIA OptiX
  • Blender
        Blend File: Pabellon Barcelona - Compute: CUDA
  • Blender
        Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX
  • Blender
        Blend File: Barbershop - Compute: CUDA
  • Blender
        Blend File: Barbershop - Compute: NVIDIA OptiX
  • Caffe
        Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 100
  • Caffe
        Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 200
  • Caffe
        Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 1000
  • Caffe
        Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 100
  • Caffe
        Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 200
  • Caffe
        Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 1000
  • VkFFT
  • NCNN
        Target: Vulkan GPU
  • RealSR-NCNN
        Scale: 4x - TAA: Yes
  • RealSR-NCNN
        Scale: 4x - TAA: No

Revision History

pts/nvidia-gpu-compute-1.0.6     20 Oct 2020 14:12 EDT
Fix for NCNN argument.

pts/nvidia-gpu-compute-1.0.5     20 Oct 2020 14:08 EDT
Add waifu2x-ncnn test profile.

pts/nvidia-gpu-compute-1.0.4     07 Oct 2020 07:05 EDT
Add NCNN and RealSR-NCNN to suite.

pts/nvidia-gpu-compute-1.0.3     04 Oct 2020 13:03 EDT
Add latest tests to suite.

pts/nvidia-gpu-compute-1.0.2     17 Sep 2020 18:25 EDT
Add Hashcat

pts/nvidia-gpu-compute-1.0.1     13 Sep 2020 18:48 EDT
Add Blender tests.

pts/nvidia-gpu-compute-1.0.0     13 Sep 2020 18:43 EDT
Initial commit for showing tests known to be in good shape for NVIDIA GPU usage.


User Comments

Post A Comment