Changes between Version 16 and Version 17 of DataParallel/BenchmarkStatus


Ignore:
Timestamp:
Mar 5, 2009 1:42:46 AM (5 years ago)
Author:
chak
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • DataParallel/BenchmarkStatus

    v16 v17  
    4141|| DotP, primitives || 100M elements || 937/937 || 934/934 || 474/474 || 238/238 || 120/120 || 65/65 || 38/38 || 28/28 || 
    4242|| DotP, vectorised || 100M elements || 937/937 || 942/942 || 471/471 || 240/240 || 118/118 || 65/65 || 43/43 || 29/29 ||  
    43 || DotP, ref Haskell || 100M elements || – || || || || || || || || 
     43|| DotP, ref Haskell || 100M elements || – || 934 || 467 || 238 || 117 || 61 || 65 || 36 || 
    4444|| DotP, ref C || 100M elements || – || || || || || || || || || 
    4545|| SMVM, primitives || ?? elems, density ?? ||  || || || || || || || || 
     
    5050==== Observations regarding DotP ==== 
    5151 
    52 The benchmark scales nicely up to the maximum number of hardware threads.  Memory latency is largely covered by excess parallelism. 
     52The benchmark scales nicely up to the maximum number of hardware threads.  Memory latency is largely covered by excess parallelism.  It is unclear why the Haskell reference implementation "ref Haskell" falls of at 32 and 64 threads. 
    5353 
    5454----