Changes between Version 49 and Version 50 of DataParallel/BenchmarkStatus
 Timestamp:
 Dec 2, 2010 2:48:47 AM (3 years ago)
Legend:
 Unmodified
 Added
 Removed
 Modified

DataParallel/BenchmarkStatus
v49 v50 8 8 9 9 == Key == 10 <project>.<benchmark name>.<version>.<parallelism>.<threads>10 <project>.<benchmark>.<version>.<parallelism>.<threads> 11 11 12 12 Project … … 29 29 30 30 [http://darcs.haskell.org/packages/dph/dphexamples/imaginary/SumSquares/ SumSquares]:: 31 Computes the sum of the squares from 1 to N using `Int`. N = 10 M.31 Computes the sum of the squares from 1 to N using `Int`. N = 100M. 32 32 33 33  '''name'''  '''runtime'''  '''speedup'''  '''notes''' … … 38 38  dph.sumsq.vectorised.par.N4  111ms  3.63   39 39 40 '''Summary''': fine, though we should run a sequential C version as well. 40 '''Summary''': fine[[br]] 41 '''Todo''': Add the sequential C version. 41 42 42 43 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 highlevel DPH program transformed by GHC's vectoriser. In addition to these two DPH variants of the dot product, we also have two nonDPH 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/dphexamples/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. 45 58 46 59 [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 highlevel 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. 48 63 49 64 [[br]]