DOI

  • A Gonzáles-Escribano
  • AJC van Gemund
  • V Cardeñoso-Payo
  • R Portales-Fernández
  • JA Caminero-Granja
Nested-parallel programming models, where the task graph associated to a computation is series-parallel, present good analysis properties that can be exploited for scheduling, cost estimation or automatic mapping to different architectures. In this work we present a preliminary framework approach to exploit some of these advantages. In our framework we reconstruct an application task graph from a high-level specification, where no scheduling or communication details are yet expressed. The obtained synchronization structure determines which mapping modules or back-ends are used to port the application to an specific platform. The first results obtained with our prototype show that even simple balancing techniques for irregular scientific applications may be easily integrated in this nested-parallel framework, to obtain efficient implementations from high-level and portable specifications. Topic: Parallel and Distributed Computing.
Original languageUndefined/Unknown
Title of host publicationHigh performance computing for computational science - VECPAR 2004
EditorsM Daydé, JJ Dongarra, V Hernández, JMLM Palma
Place of PublicationBerlin
PublisherSpringer
Pages541-555
Number of pages15
ISBN (Print)3-540-25424-2
DOIs
Publication statusPublished - 2005
Event6th International Conference, Valencia, Spain - Berlin
Duration: 28 Jun 200430 Jun 2004

Publication series

Name
PublisherSpringer
NameLecture Notes in Computer Science
Volume3402
ISSN (Print)0302-9743

Conference

Conference6th International Conference, Valencia, Spain
Period28/06/0430/06/04

    Research areas

  • academic journal papers, ZX CWTS JFIS < 1.00

ID: 3901319