A framework for adaptive collective communications for heterogeneous hierarchical computing systems

Collective communication operations are widely used in MPI applications and play an important role in their performance. However, the network heterogeneity inherent to grid environments represent a great challenge to develop efficient high performance computing applications. In this work we propose...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of computer and system sciences Jg. 74; H. 6; S. 1082 - 1093
Hauptverfasser: Steffenel, Luiz Angelo, Mounié, Grégory
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Inc 2008
Elsevier
Schlagworte:
Adaptive techniques
Collective communication
Computer Science
Distributed, Parallel, and Cluster Computing
Grid computing
MPI
Performance modeling
Polyalgorithms
Performance modeling
Grid computing
Adaptive techniques
Collective communication
Polyalgorithms
MPI
ISSN:0022-0000, 1090-2724
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Collective communication operations are widely used in MPI applications and play an important role in their performance. However, the network heterogeneity inherent to grid environments represent a great challenge to develop efficient high performance computing applications. In this work we propose a generic framework based on communication models and adaptive techniques for dealing with collective communication patterns on grid platforms. Toward this goal, we address the hierarchical organization of the grid, selecting the most efficient communication algorithms at each network level. Our framework is also adaptive to grid load dynamics since it considers transient network characteristics for dividing the nodes into clusters. Our experiments with the broadcast operation on a real-grid setup indicate that an adaptive framework allows significant performance improvements on MPI collective communications.
ISSN:0022-0000
1090-2724
DOI:10.1016/j.jcss.2007.07.010