Combining Hilbert SFC and bruijn graphs for searching computing markets in a P2P system

  • Authors:
  • Damia Castellà;Hector Blanco;Francesc Giné;Francesc Solsona

  • Affiliations:
  • Dept. of Computer Science, University of Lleida, Lleida, Spain;Dept. of Computer Science, University of Lleida, Lleida, Spain;Dept. of Computer Science, University of Lleida, Lleida, Spain;Dept. of Computer Science, University of Lleida, Lleida, Spain

  • Venue:
  • EuroPar'10 Proceedings of the 16th international Euro-Par conference on Parallel processing: Part I
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes an efficient and scalable computational resource discovery overlay orientated towards P2P computing. Our proposal gathers the peers into markets according to their computational resources. Each market is arranged in an N-tree and the trees are linked by a Bruijn graph. The tree topology allows efficient searching of available resources in a specific market, while Bruijn provides good scalability because search complexity does not depend on the number of markets. A Hilbert function is used to arrange markets in one ordered and monodimensional space. This way, the proposed architecture exploits the Bruijn and N-tree topologies together with the Hilbert function. A look-up query mechanism for simple and multiple queries with a low algorithmic cost is also introduced over this architecture. The performance of our proposal was analysed by means of simulation in relation to the widely used Chord overlay with the case of simple queries, and the Baton algorithm with the case of range queries. Furthermore, a large number of experiments demonstrate the proper behaviour of the system. The results obtained reveal the competitiveness of our proposals.