Basic HTML version of Foils prepared August 7 98

Foil 28 Implementation of INDEPENDENT

From DoD HPF Training -- 3. Parallel Constructs in HPF DoD Training and Others -- 1995-98. by Chuck Koelbel -- Rice University
1 Virtually identical to FORALL
2 Use owner-computes rule to partition computation
  • See implementation of DISTRIBUTE
3 Semantics of the construct allow parallelism
  • All iterations can execute in parallel
4 Data need only be moved at synchronization points
  • Before and after the loop
5 No further dependence analysis needed
6 Warnings about functions still apply
  • Possibly worse, due to more pathological cases
in Table To:

© Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Sun Aug 9 1998