Adaptive expression based routing protocol for P2P systems

  • Authors:
  • Imran Rao;Aaron Harwood;Shanika Karunasekera

  • Affiliations:
  • NICTA Victoria Labs, CSSE, The University of Melbourne Australia;NICTA Victoria Labs, CSSE, The University of Melbourne Australia;NICTA Victoria Labs, CSSE, The University of Melbourne Australia

  • Venue:
  • OTM'07 Proceedings of the 2007 OTM Confederated international conference on On the move to meaningful internet systems - Volume Part II
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

There is an emerging trend of using P2P systems for computational and data intensive tasks, such as online collaborations, distributed database applications and message passing interface (MPI) algorithms. For better resource utilization in these and similar applications, it is necessary to discover the resource capabilities of the collaborating peers. Another requirement of these applications is to locate the optimal resource based on a search criteria, e.g., to seek a resource with earliest execution time. We define this kind of routing as expression based routing. However, in the absence of the centralized controlling node, tracking the capabilities of the participating peers is very difficult. Moreover peers join and leave the system dynamically and, thus, makes the discovery of a desired resource even more complex. In this paper we investigate a novel algorithm for expression based routing in a P2P system. It evaluates peer u such that lu (the value of u) is minimum where e.g. lu is load of the peer u. Our contribution includes a detailed algorithm to search for the least loaded peer in the system. We analyze the accuracy and cost of our proposed algorithm through detailed simulations.