Changes between Version 9 and Version 10 of DataParallel/WorkPlan


Ignore:
Timestamp:
Jan 30, 2009 5:53:07 AM (6 years ago)
Author:
chak
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • DataParallel/WorkPlan

    v9 v10  
    11 
    22== Work plan for implementing Data Parallel Haskell == 
     3 
     4=== Milestones === 
     5 
     6 0. '''DUE 6 March.''' Solve major performance and scalability problems for our current benchmarks.  Try to get a performance advantage over plain Haskell on !LimitingFactor (i.e., 8 cores). 
     7 0. '''DUE 30 March.''' Presentation for ''Microsoft External Research Symposium''. 
    38 
    49=== Task assignments === 
    510 
    611 ''Roman'':: 
    7    '''Replicate''' & '''Recycling''' 
     12   '''Replicate''' & #2984 & '''Recycling''' 
    813   – status: partly implemented, but still needs serious work 
     14   * To use the special representation of task '''Replicate''' most effectively, we would ''again'' need different views on arrays together with a cost function and optimisation rules taking the cost function into account.  That requires a lot of work! 
     15   * We decided that, for the moment, Roman will first try to integrate the replication representation directly and see how far that gets us.  Maybe it helps at least with some examples and gives us something somewhat usable more quickly. 
    916 
    1017 ''Simon'':: 
    11    '''!CoreToStg''' & '''Code blow up''' 
     18   '''Code blow up''' 
    1219   – status: unknown 
    1320 
    1421 ''Gabi'':: 
    15    '''Hierarchical matrix representation''' 
     22   '''Hierarchical matrix representation''' & '''Benchmark status''' 
    1623   – status: just started 
    1724 
     
    2431Category: ''Bugs'' 
    2532 
    26  * '''!CoreToStg''': Compiling package dph with the HEAD currently results in `ASSERT failed! file stgSyn/CoreToStg.lhs line 239` (with a DEBUG compiler). 
     33 * #2984 
    2734 
    2835Category: ''Efficiency'' (improve scalability and/or baseline performance of generated code): 
     
    5663 * '''Hierarchical matrix representation:''' Sparse matrices can be space-efficiently represented by recursively decomposing them into four quadrants.  Decomposition stops if a quadrant is smaller than a threshold or contains only zeros.  Multiplication of such matrices is straight forward using Strassen's divide-and-conquer scheme, which is popular for parallel implementations.  Other operations, such as transposing a matrix, can also be efficiently implemented.  The plan is to experiment with the implementation of some BLAS routines using this representation. 
    5764 
     65 * '''Benchmark status:''' Update and complete [wiki:DataParallel/BenchmarkStatus]. 
     66 
    5867 * '''N-body:''' Get a fully vectorised n-body code to run and scale well on !LimitingFactor. 
     68 
     69---- 
     70 
     71=== Done === 
     72 
     73 * '''!CoreToStg''': Compiling package dph with the HEAD currently results in `ASSERT failed! file stgSyn/CoreToStg.lhs line 239` (with a DEBUG compiler). 
     74