cory-master/BabelStream
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
80 Commits 1 Branch 0 Tags 2.5 MiB
C++ 52.4%
Fortran 10.1%
CMake 9.7%
Java 7.3%
Shell 5.5%
Other 15%
70ccd22919
Go to file
Tom Deakin 70ccd22919 Free CUDA buffers
2015-07-31 13:59:34 +01:00
CL new files 2015-07-28 11:37:50 +01:00
results Update S9150 result with new default array size 2015-07-31 11:22:35 +01:00
.gitignore new files 2015-07-28 11:37:50 +01:00
common.cpp Use binary power of two for default array size 2015-07-31 11:20:39 +01:00
common.h Bump version to 0.9 2015-07-30 13:57:29 +01:00
cuda-stream.cu Free CUDA buffers 2015-07-31 13:59:34 +01:00
LICENSE Update License 2015-07-30 13:57:00 +01:00
Makefile Remove cuda check in main line Makefile - only needs to be in CUDA part 2015-07-28 12:09:23 +01:00
ocl-stream-kernels.cl Update License 2015-07-30 13:57:00 +01:00
ocl-stream.cpp Update License 2015-07-30 13:57:00 +01:00
README.md Update README 2015-07-16 18:23:28 +01:00

README.md

GPU-STREAM

Measure memory transfer rates to/from global device memory on GPUs. This benchmark is similar in spirit, and based on, the STREAM benchmark [1] for CPUs.

Unlike other GPU memory bandwidth benchmarks this does not include the PCIe transfer time.

Usage

Build the OpenCL and CUDA binaries with make

Run the OpenCL version with ./gpu-stream-ocl and the CUDA version with ./gpu-stream-cuda

Results

Sample results can be found in the results subdirectory. If you would like to submit updated results, please submit a Pull Request.

[1]: McCalpin, John D., 1995: "Memory Bandwidth and Machine Balance in Current High Performance Computers", IEEE Computer Society Technical Committee on Computer Architecture (TCCA) Newsletter, December 1995.