Skip to content
Snippets Groups Projects
Select Git revision
0 results
Search by author
  • Any Author
0 authors
  1. Aug 09, 2015
  3. Jun 30, 2015
  5. Jun 21, 2015
  7. Jul 17, 2014
  9. Jun 15, 2014
  11. May 29, 2014
  13. Apr 29, 2014
  15. Oct 17, 2013
  17. Sep 16, 2013
  19. Aug 17, 2013
  21. Jul 29, 2013
  23. Jul 07, 2013
  25. Jul 06, 2013
  27. Jun 10, 2013
  29. Mar 20, 2013
  31. 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
  33. Oct 06, 2012
  35. Aug 11, 2012
  37. Jul 30, 2012
  39. Jul 25, 2012
  41. Jun 18, 2012
  43. Dec 07, 2011
  45. Aug 22, 2011
  47. Aug 08, 2011
  49. Jul 04, 2011
  51. May 07, 2011
  53. Apr 29, 2011
  55. Apr 17, 2011
  57. Feb 25, 2011
  59. Feb 18, 2011
  61. Feb 17, 2011
  63. Jan 19, 2011
  65. Jan 18, 2011
  67. Jan 01, 2011
  69. Dec 15, 2010