https://github.com/gante/opencl-parallel-reduction

OpenCL code for the Parallel Reduction, with good results on FPGAs
https://github.com/gante/opencl-parallel-reduction

Last synced: 23 days ago
JSON representation

OpenCL code for the Parallel Reduction, with good results on FPGAs

• Host: GitHub
• URL: https://github.com/gante/opencl-parallel-reduction
• Owner: gante
• License: mit
• Created: 2018-01-29T21:46:39.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-01-30T10:44:42.000Z (almost 7 years ago)
• Last Synced: 2024-12-03T15:49:16.637Z (about 1 month ago)
• Language: C++
• Homepage:
• Size: 15.6 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# OpenCL-Parallel-Reduction

OpenCL code for the Parallel Reduction, able to run on FPGAs (with good results on GPUs, only requires minimal code changes for GPUs)

Based on: http://developer.download.nvidia.com/compute/cuda/1.1-Beta/x86_website/projects/reduction/doc/reduction.pdf

Stage by stage optimization of a memory bound problem. 

//////////////////////////////////////////////////////

version 8 vs version 1:

Speedup for GTX860M - 12.85x @ BW=58.46GB/s

Speedup for DE5-Net (Stratix V 5SGXA7) - 30.84x @ BW = 4.35GB/s