Changes between Version 49 and Version 50 of DataParallel/BenchmarkStatus


Ignore:
Timestamp:
Dec 2, 2010 2:48:47 AM (3 years ago)
Author:
benl
Comment:

Update numbers

Legend:

Unmodified
Added
Removed
Modified
  • DataParallel/BenchmarkStatus

    v49 v50  
    88 
    99== Key == 
    10 <project>.<benchmark-name>.<version>.<parallelism>.<threads> 
     10<project>.<benchmark>.<version>.<parallelism>.<threads> 
    1111 
    1212Project 
     
    2929 
    3030 [http://darcs.haskell.org/packages/dph/dph-examples/imaginary/SumSquares/ SumSquares]:: 
    31   Computes the sum of the squares from 1 to N using `Int`.  N = 10M. 
     31  Computes the sum of the squares from 1 to N using `Int`.  N = 100M. 
    3232 
    3333  || '''name''' || '''runtime''' || '''speedup''' || '''notes''' 
     
    3838  || dph.sumsq.vectorised.par.N4 || 111ms || 3.63 || || 
    3939 
    40   '''Summary''': fine, though we should run a sequential C version as well. 
     40  '''Summary''': fine[[br]] 
     41  '''Todo''': Add the sequential C version. 
    4142 
    4243 
    43  [http://darcs.haskell.org/packages/dph/examples/dotp/ DotP]:: 
    44   Computes the dot product of two vectors of `Double`s.  There are two variants of this program: (1) "primitives" is directly coded against the array primitives from package dph and (2) "vectorised" is a high-level DPH program transformed by GHC's vectoriser.  In addition to these two DPH variants of the dot product, we also have two non-DPH reference implementations: (a) "ref Haskell" is a Haskell program using imperative, unboxed arrays and and (b) "ref C" is a C implementation using pthreads. 
     44 [http://darcs.haskell.org/packages/dph/dph-examples/imaginary/DotProduct DotProduct]:: 
     45  Computes the dot product of two vectors of `Double`s. N=10M. 
     46 
     47  || '''name''' || '''runtime''' || '''speedup''' || '''notes''' 
     48  || dph.sumsq.vector.seq.N4 ||  68ms || 1 || || 
     49  || dph.sumsq.vectorised.seq.N4 || 58ms || 1.17 || A || 
     50  || dph.sumsq.vectorised.par.N1 || 55ms || 1.24 || || 
     51  || dph.sumsq.vectorised.par.N2 || 33ms || 2.06 || || 
     52  || dph.sumsq.vectorised.par.N4 || 25ms || 2.72 || || 
     53  
     54  A: The vectorised version is faster than with Data.Vector. Why was this? 
     55 
     56  '''Summary''': fine.[[br]] 
     57  '''Todo''': Add the sequential C version. 
    4558 
    4659 [http://darcs.haskell.org/packages/dph/examples/smvm/ SMVM]:: 
    47   Multiplies a dense vector with a sparse matrix represented in the ''compressed sparse row format (CSR).''  There are three variants of this program: (1) "primitives" is directly coded against the array primitives from package dph and (2) "vectorised" is a high-level DPH program transformed by GHC's vectoriser.  As a reference implementation, we have a sequential C program denoted by "ref C". 
     60  Multiplies a dense vector with a sparse matrix represented in the ''compressed sparse row format (CSR).''   
     61 
     62  '''Todo''': Add this to the nightly run. 
    4863 
    4964[[br]]