Analyzing Software Component Graphs of Grid Middleware: Hint to Performance Improvement

  • Authors:
  • Pingpeng Yuan;Hai Jin;Kang Deng;Qingcha Chen

  • Affiliations:
  • Service Computing Technology and System Lab Cluster and Grid Computing Lab, Huazhong University of Science and Technology, Wuhan, China 430074;Service Computing Technology and System Lab Cluster and Grid Computing Lab, Huazhong University of Science and Technology, Wuhan, China 430074;Service Computing Technology and System Lab Cluster and Grid Computing Lab, Huazhong University of Science and Technology, Wuhan, China 430074;Service Computing Technology and System Lab Cluster and Grid Computing Lab, Huazhong University of Science and Technology, Wuhan, China 430074

  • Venue:
  • ICA3PP '08 Proceedings of the 8th international conference on Algorithms and Architectures for Parallel Processing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Grid middleware is a kind of important service management tool and is composed of different interaction units. The units of a grid middleware do not interact in random ways and are very well connected. As shown in this paper, irrespective of the specific features of each grid middleware analyzed, the final outcome of grid middleware is a small world, hierarchical component diagram with well-defined statistical properties. These measurements of network are largely independent of the particular the application and indicated key execution path or key elements. Therefore, analyzing grid middleware structure can show a roadmap to tune performance of grid middleware. Based on analysis of those diagrams, the key components of grid middleware are outlined.