Design Issues of a Novel Toolkit for Parallel Application Performance Monitoring and Analysis in Cluster and Grid Environments

  • Authors:
  • Kuan-Ching Li;Hsiao-Hsi Wang;Chiou-Nan Chen;Chun-Chieh Liu;Chia-Fu Chang;Chia-Wen Hsu;Sheng-Shiang Hung

  • Affiliations:
  • Providence University Shalu, Taiwan;Providence University Shalu, Taiwan;National Tsing Hua University ,Taiwan;Providence University Shalu, Taiwan;Providence University Shalu, Taiwan;Providence University Shalu, Taiwan;Providence University Shalu, Taiwan

  • Venue:
  • ISPAN '05 Proceedings of the 8th International Symposium on Parallel Architectures,Algorithms and Networks
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The low-cost and availability of network of workstations have made them attractive solution for high performance computing. Striking progress of network technology is enabling high-performance global computing, with the utilization of cluster and grid technologies, in which computational and data resources in a local or wide area network are transparently employed to solve large-scale problems. In this paper, we present implementation and design rationale of Visuel* toolkit for performance measurement and analysis of parallel applications in cluster and grid environments. Most of performance visualization tools available today for highperformance platforms show solely system performance data (e.g., CPU load, memory usage, network bandwidth, server average load), and thus, being suitable for visualization of computing platform system activities. The Visuel toolkit is webbased interface designed to show performance activities of all computing nodes of a cluster or grid computing platform involved in the execution of a parallel application, such as CPU load level and memory usage of each computing node. In addition, this toolkit is able to display comparative performance data visualizations generated from a number of executions of an application under investigation, analyzing the performance of different implementations. Evaluations using this toolkit show that it outperforms in easing the process of investigation and implementation of parallel applications, in effective way.