Skip to content
Snippets Groups Projects
  1. Apr 29, 2014
  2. Oct 17, 2013
  3. Sep 16, 2013
  4. Aug 17, 2013
  5. Jul 29, 2013
  6. Jul 07, 2013
  7. Jul 06, 2013
  8. Jun 10, 2013
  9. Mar 20, 2013
  10. 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
  11. Oct 06, 2012
  12. Aug 11, 2012
  13. Jul 30, 2012
  14. Jul 25, 2012
  15. Jun 18, 2012
  16. Dec 07, 2011
  17. Aug 22, 2011
  18. Aug 08, 2011
  19. Jul 04, 2011
  20. May 07, 2011
  21. Apr 29, 2011
  22. Apr 17, 2011
  23. Feb 25, 2011
  24. Feb 18, 2011
  25. Feb 17, 2011
  26. Jan 19, 2011
  27. Jan 18, 2011
  28. Jan 01, 2011
  29. Dec 15, 2010
  30. Oct 31, 2010
  31. Oct 20, 2010
  32. Sep 17, 2010
  33. Aug 06, 2010
  34. Jul 20, 2010
  35. Jul 16, 2010
Loading