Characterizing and Evaluating Desktop Grids: An Empirical Study

Overview

Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been little in-sight into the detailed temporal structure of CPU availability of desktop grid resources. Yet, this structure is critical to characterize the utility of desktop grid platforms for both task parallel and even data parallel applications. This paper addresses the following questions: What are the temporal characteristics of desktop CPU availability in an enterprise setting? How do these characteristics affect the utility of desktop grids? Based on these characteristics, can one construct a model of server "Equivalents" for the desk-top grids, which can be used to predict application performance?