Predicting the Resource Requirements of a Job Submission

Arshad Ali, Ashiq Anjum, Julian Bunn, Richard Cavanaugh, Frank van Lingen, Richard McClatchey, Muhammad Atif Mehmood, Harvey Newman, Conrad Steenberg, Michael Thomas and Ian Willers (2004) Predicting the Resource Requirements of a Job Submission. In: Computing in High Energy Physics, 2004, Interlaken, Switzerland. [CaltechCACR:2004.210]

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

Grid computing aims to provide an infrastructure for distributed problem solving in dynamic virtual organizations. It is gaining interest among many scientific disciplines as well as the industrial community. However, current grid solutions still require highly trained programmers with expertise in networking, high-performance computing, and operating systems. One of the big issues in full-scale usage of a grid is matching the resource requirements of job submission to the resources available on the grid. Resource brokers and job schedulers must make estimates of the resource usage of job submissions in order to ensure efficient use of grid resources. We prop ose a prediction engine that will operate as part of a grid scheduler. This prediction engine will provide estimates of the resources required by job submission based upon historical information. This paper presents the need for such a prediction engine and discusses two approaches for history based estimation.

EPrint Type:	Conference or Workshop Item (Paper)
Subjects:	All Records
ID Code:	55
Deposited By:	Sarah M. Emery
Deposited On:	16 November 2004
Record Number:	CaltechCACR:2004.210
Official Persistent URL:	http://resolver.caltech.edu/CaltechCACR:2004.210
Usage Policy:	You are granted permission for individual, educational, research and non-commercial reproduction, distribution, display and performance of this work in any format.

Archive Staff Only: edit this record