ScienceDirect - Parallel Computing : Mapping subtasks with multiple versions on an ad hoc grid

Parallel Computing
Volume 31, Issue 7, July 2005, Pages 671-690
Heterogeneous Computing

Font Size:

Abstract - selected

E-mail Article

Add to my Quick Links

Bookmark and share in 2collab (opens in new window)

Request permission to reuse this article

Cited By in Scopus (3)

Related Articles in ScienceDirect

Static allocation of resources to communicating subtask...
Journal of Parallel and Distributed Computing

On the optimal order of M machines in tandem
Operations Research Letters

A practical model of Heineken's bottle filling line wit...
European Journal of Operational Research

Optimal preemptive semi-online scheduling to minimize m...
Operations Research Letters

Techniques for mapping tasks to machines in heterogeneo...
Journal of Systems Architecture

View More Related Articles

View Record in Scopus

doi:10.1016/j.parco.2005.04.003

Mapping subtasks with multiple versions on an ad hoc grid

S. Shivle^a^,, P. Sugavanam^a^,, H.J. Siegel^a^,^b^,^,, A.A. Maciejewski^a^,, T. Banka^a^,, K. Chindam^a^,, S. Dussinger^c^,, A. Kutruff^a^,, P. Penumarthy^a^,, P. Pichumani^a^,, P. Satyasekaran^a^,, D. Sendek^a^,, J. Smith^d^,, J. Sousa^c^,, J. Sridharan^a^, and J. Velazco^e^,

^aColorado State University, Department of Electrical and Computer Engineering, Fort Collins CO 80523-1373, USA ^bColorado State University, Department of Computer Science, Fort Collins, CO 80523, USA ^cHP Technologies, Fort Collins, CO 80528-9544, USA ^dIBM Corporation, Boulder, CO 80301, USA ^eStelex, San Juan, Puerto Rico

Received 31 October 2004;

revised 15 March 2005;

accepted 15 April 2005.

Available online 16 June 2005.

References and further reading may be available for this article. To view references and further reading you must purchase this article.

Abstract

An ad hoc grid is a heterogeneous computing system composed of mobile devices. Each computing resource is constrained in battery energy. The problem being studied is to assign statically computing resources to the subtasks of an application that has an execution time constraint, when the resources are oversubscribed. All subtasks must be executed; to accommodate this in an oversubscribed environment, each subtask has two versions: the primary or full version, and the secondary or degraded version. The secondary version utilizes only 10% of the resources that the primary version requires, and produces only 10% of the data output for the subsequent children subtasks. Thus, the degraded version (secondary version) represents a reduced capability of lesser overall value, while consuming fewer resources. The goal is to assign resources so that the application meets an execution time constraint and the battery energy constraint while minimizing the number of degraded versions used. Five resource allocation heuristics to derive near-optimal solutions to this problem are presented, evaluated, and compared.

Keywords: Ad hoc grid; Communication scheduling; Mapping; Resource allocation; Task scheduling