Results 1 - 3 of 3

Intra node parallelization of MPI programs with OpenMP

by Franck Cappello, Olivier Richard , 1998
"... The availability of multiprocessors and high performance networks offer the opportunity to construct CLUMPs (Cluster of Multiprocessors) and use them as paxallel computing platforms. The main distinctive feature of the CLUMP axchitecture over the usual paxallel computers is its hybrid memory model ..."
Abstract - Cited by 3 (0 self) - Add to MetaCart
of shaxed memory paxallel programs. Second, it investigates the method to transform MPI paxallel programs in order to execute them on a CLUMP. Third, it presents the performance evaluation of this method applied on the NAS paxallel benchmaxks executed on a cluster of biprocessor PCs.

Effects of Ordering Strategies and Programming . . .

by Leonid Oliker, Xiaoye Li, Parry Husbands, Rupak Biswas - IN SIAM REVIEW , 2002
"... The conjugate gradient (CG) a#G).K--1; is perha#8 the best-knownitera#55 e technique for solvingspa#in linea# systemstha# a#a symmetrica#m positive definite.For systemstha# a#a ill conditioned, it is oftennecessa#W to usea preconditioning technique.In thispa# er, we investiga#. the e#ects of va#KK61 ..."
Abstract - Add to MetaCart
#i distributedsha#tri memory systems,ca# he reusema y be more importa# ttha# reducing communica#WYK. it is possible to a# hieve messa#6.x6WW--1. performa#6W usingsha#.6;SW8.x6 constructs through ca#ough da#o orderinga#d distribution,a#i a hybrid MPI+OpenMPpa#P.688 increa#8W progra#WK.x complexity with little

Automatic Parallelization by Pattern-Matching

by Christoph W. Kebler, Wolfgang J. Paul, Fachbereich Informatik
"... Abstract. We present the top-down design of a new system which per-forms automatic parallelization of numerical Fortran 77 or C source pro-grams for execution on distributed-memory message- passing multi-processors such as e.g. the INTEL iPSC860 or the TMC CM-5. The key idea is a high-level pattern- ..."
Abstract - Add to MetaCart
Abstract. We present the top-down design of a new system which per-forms automatic parallelization of numerical Fortran 77 or C source pro-grams for execution on distributed-memory message- passing multi-processors such as e.g. the INTEL iPSC860 or the TMC CM-5. The key idea is a high-level pattern
Results 1 - 3 of 3