Advertisement

Efficient Top-k Query Processing in P2P Network

  • Yingjie He
  • Yanfeng Shu
  • Shan Wang
  • Xiaoyong Du
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3180)

Abstract

In this paper, we examine how to efficiently process top-k queries in pure P2P network. First, we propose a distributed algorithm to evaluate a top-k query in a hierarchical way. Each peer does its local top-k query, and top k results from different peers are merged hierarchically from bottom to the top (the root peer is the query initiator). Ranking and merging of results are distributed across the peers to exploit the computing resources in the network. Second, to improve performance, we maintain histograms at each peer according to the top k results returned by the peers. The histograms are used to estimate the possible upper bound scores of peers so that a query only needs to be forwarded to the most promising neighbouring peers. Our experimental study shows that the top-k query algorithm improves the query effectiveness, while the use of histograms enhances the query efficiency.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Baeza-Yates, R., Ribeiro-Neto, B.: Modern information retrieval. Addison-Wesley, Reading (1999)Google Scholar
  2. 2.
    Callan, J., Croft, W., Harding, S.: The inquery retrieval system. In: Proc. of DEXA (1992)Google Scholar
  3. 3.
    Callan, J., Lu, Z., Croft, W.: Searching distributed collections with inference networks. In: Proc. of ACM SIGIR (1995)Google Scholar
  4. 4.
    Cuenca-Acuna, F., Nguyen, T.: Text-based content search and retrieval in ad hoc p2p communities. Technical report DCS-TR-483, Rutgers University (2002)Google Scholar
  5. 5.
    Gnutella Development Home Page, http://gnutella.wego.com/
  6. 6.
    Gravano, L., Garcia-Molina, H., Tomasic, A.: The effectiveness of gloss for the text database discovery problem. In: Proc. of ACM SIGMOD (1994)Google Scholar
  7. 7.
    Napster Home Page, http://www.napster.com/
  8. 8.
    Palmer, C., Steffan, J.: Generating network topologies that obey power law. In: Proc. of GLOBECOM (2000)Google Scholar
  9. 9.
    Ripeanu, M.: Peer-to-peer architecture case study: Gnutella network. Technical report TR-2001-26, University of Chicago (July 2001)Google Scholar
  10. 10.
    Stoica, I., Morris, R., Karger, D., Kaashoek, F., Balakrishnan, H.: Chord: A scalable peer-to-peer lookup service for internet applications. In: Proc. of ACM SIGCOMM (2001)Google Scholar
  11. 11.
    Tang, C., Xu, Z., Mahalingam, M.: Peersearch: efficient information retrieval in peer-to-peer networks. Technical report HPL-2002-198, HP Labs (2002)Google Scholar
  12. 12.
    Yu, C., Philip, G., Meng, W.: Distributed top-n query processing with possibly uncooperative local systems. In: Proc. of VLDB (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Yingjie He
    • 1
  • Yanfeng Shu
    • 2
  • Shan Wang
    • 1
  • Xiaoyong Du
    • 1
  1. 1.Information SchoolRenmin University of ChinaBeijingChina
  2. 2.Dept of Computer ScienceNational University of SingaporeSingapore

Personalised recommendations