A dynamic and scalable performance monitoring toolkit for cluster and grid environments

Abstract

The low-cost and availability of network of workstations have made them an attractive solution for high performance computing. Striking progress of network technology is enabling high-performance global computing in which large sets of diverse and geographically distributed resources are transparently and coordinately employed to solve large-scale problems. In this paper, Visuel toolkit for performance measurement and analysis of parallel applications in cluster and grid environments is presented. It is a web-based interface designed not only to provide dynamic monitoring view and management information, but also performance activity of all computing nodes of computing platform involved in the execution of a parallel application. Evaluations using this toolkit show the effectiveness in easing the process of investigation and implementation of parallel applications.