VO-Level Performance Analysis of gLite Grids

Abstract

Business oriented grids call for effective Quality of Service strategies able to adapt to different user requirements. Performance analyses and predictions with respect to different load conditions or management policies are essential instruments to define such strategies. In this paper, we present a non-Markovian Stochastic Petri Net model that allows to conduct performance analyses of Grid systems focusing on aspects related to the Virtual Organization as a whole. Different job allocation techniques will be evaluated with respect to both user and provider point-of-views. We will also investigate the influence of different information update policies on the accuracy of the allocation schemes, highlighting the costs/benefits in terms of job waiting time, service availability, and system utilization.