Skip to content
Snippets Groups Projects
  1. Jul 17, 2014
  2. Jun 15, 2014
  3. May 29, 2014
  4. Apr 29, 2014
  5. Oct 17, 2013
  6. Sep 16, 2013
  7. Aug 17, 2013
  8. Jul 29, 2013
  9. Jul 07, 2013
  10. Jul 06, 2013
  11. Jun 10, 2013
  12. Mar 20, 2013
  13. Oct 24, 2012
    • dhj's avatar
      Updated benchmark to use equivalent work loads for cpu and gpu. · fa833504
      dhj authored
      * Numpy does element-wise operations by default.  Updated the cpu
        operation to use pure numpy.
      
      * Eliminated the loop which is not necessary to demonstrate
        parallelism on array operations.
      
      * Made the number of workers explicit rather than gpu chosen,
        through local_size variable passed to kernel execution.
      
      * Increased to ~8 million data points to more clearly demonstrate
        the difference between cpu and gpu based computations.
      fa833504
  14. Oct 06, 2012
  15. Aug 11, 2012
  16. Jul 30, 2012
  17. Jul 25, 2012
  18. Jun 18, 2012
  19. Dec 07, 2011
  20. Aug 22, 2011
  21. Aug 08, 2011
  22. Jul 04, 2011
  23. May 07, 2011
  24. Apr 29, 2011
  25. Apr 17, 2011
  26. Feb 25, 2011
  27. Feb 18, 2011
  28. Feb 17, 2011
  29. Jan 19, 2011
  30. Jan 18, 2011
  31. Jan 01, 2011
  32. Dec 15, 2010
  33. Oct 31, 2010
  34. Oct 20, 2010
Loading