Skip to Main content Skip to Navigation
Reports

Load-Balancing Scatter Operations for Grid Computing

Stéphane Genaud 1 Arnaud Giersch Frédéric Vivien
1 REMAP - Regularity and massive parallel computing
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : We present solutions to statically load-balance scatter operations in parallel codes run on grids. Our load-balancing strategy is based on the modification of the data distributions used in scatter operations. We study the replacement of scatter operations with parameterized scatters, allowing custom distributions of data. The paper presents: 1) a general algorithm which finds an optimal distribution of data across processors; 2) a quicker guaranteed heuristic relying on hypotheses on communications and computations; 3) a policy on the ordering of the processors. Experimental results with an MPI scientific code illustrate the benefits obtained from our load-balancing.
Document type :
Reports
Complete list of metadata

https://hal.inria.fr/inria-00071816
Contributor : Rapport de Recherche Inria <>
Submitted on : Tuesday, May 23, 2006 - 6:52:50 PM
Last modification on : Saturday, September 11, 2021 - 3:17:50 AM

Identifiers

  • HAL Id : inria-00071816, version 1

Collections

Citation

Stéphane Genaud, Arnaud Giersch, Frédéric Vivien. Load-Balancing Scatter Operations for Grid Computing. [Research Report] RR-4770, LIP RR-2003-17, INRIA, LIP. 2003. ⟨inria-00071816⟩

Share

Metrics

Record views

280

Files downloads

808