A Matrix Inversion Method with YML/OmniRPC on a Large Scale Platform

Authors:
Maxime Hugues;Serge G. Petiton
Affiliations:
LIFL, University of Science and Technology of Lille, Villeneuve d'Ascq, France 59655;LIFL, University of Science and Technology of Lille, Villeneuve d'Ascq, France 59655
Venue:
High Performance Computing for Computational Science - VECPAR 2008
Year:
2008

Citing 5
Cited 0

Ninf: A Network Based Information Library for Global World-Wide Computing Infrastructure

HPCN Europe '97 Proceedings of the International Conference and Exhibition on High-Performance Computing and Networking
A Parallel Adaptive version of the Block-based Gauss-Jordan Algorithm

IPPS '99/SPDP '99 Proceedings of the 13th International Symposium on Parallel Processing and the 10th Symposium on Parallel and Distributed Processing
OmniRPC: a Grid RPC ystem for Parallel Programming in Cluster and Grid Environment

CCGRID '03 Proceedings of the 3st International Symposium on Cluster Computing and the Grid
Grid'5000: A Large Scale and Highly Reconfigurable Grid Experimental Testbed

GRID '05 Proceedings of the 6th IEEE/ACM International Workshop on Grid Computing
Workflow Global Computing with YML

GRID '06 Proceedings of the 7th IEEE/ACM International Conference on Grid Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

YML is a dedicated framework to develop and run parallel applications over a large scale middleware. This framework makes easier the use of a grid and provides a high level programming tool. It is independent from middlewares and users are not in charge to manage communications. In consequence, it introduces a new level of communications and it generates an overhead. In this paper, we proposed to show the overhead of YML is tolerable in comparison to a direct use of a middleware. This is based on a matrix inversion method and a large scale platform, Grid'5000.