Pushpa Publishing House

Journal Menu

Content

Volume 17 (2018)

Volume 16 (2016)

Volume 15 (2015)

Volume 14 (2015)

Volume 13 (2014)

Volume 12 (2014)

Volume 11 (2013)

Volume 10 (2013)

Volume 9 (2012)

Volume 8 (2012)

Volume 7 (2011)

Volume 6 (2011)

	Volume 6, Issue 2 Pg 105 - 204 (May 2011)
	Volume 6, Issue 1 Pg 1 - 104 (February 2011)

Volume 5 (2010)

Volume 4 (2010)

Volume 3 (2009)

Volume 2 (2008)

Volume 1 (2007)

Advances in Computer Science and Engineering

Advances in Computer Science and Engineering
Volume 6, Issue 2, Pages 181 - 197 (May 2011)

SOME PERFORMANCE ISSUES ON LINEAR ALGEBRA ALGORITHMS IN DISTRIBUTED AND GRID COMPUTING ENVIRONMENTS
Giuliano Laccetti, Marco Lapegna, Valeria Mele and Diego Romano

Abstract:
Due to the dynamic and heterogeneous nature of grid infrastructures, scientific applications with frequent and tight synchronizations among the nodes are unable to achieve high efficiencies, so the client-server paradigm is a programming model very often used in these environments. According to this model, Data Grid applications are usually divided into independent activities that are concurrently solved by the servers. On the other hand, since many scientific applications are characterized by large collections of input data and by dependencies between the tasks, the development of efficient algorithms without unnecessary synchronizations and data transfers is a difficult task. The present work addresses the problem of implementing and assessing a strategy for efficient task scheduling and data management in case of dependencies among tasks in a numerical linear algebra problem. To this end, we used the Block Matrix Multiplication Algorithm implemented in the NetSolve distributed computing environment as case study, and we introduced some efficiency parameters to assess the algorithm.

Keywords and phrases: parallel and distributed computing, product matrix algorithms, performance analysis.

Number of Downloads: 256 | Number of Views: 813