SPANNING TREES FOR DISTRIBUTED SEARCH IN P2P SYSTEMS

Searching for data location in P2P systems is a messageintensive task. Basically, a query needs to reach all the nodes in order to find the locations of a specified data. This search involves a broadcast, and an inappropriate broadcast strategy may not only lead to inefficient data sharing, but also directly affect the scalability of the P2P system. We propose the utilization of an ad-hoc spanning tree for distributed search in P2P systems. This spanning tree is built as an overlay network. We define algorithms to maintain the ad-hoc spanning tree, and we compare the search time obtained with the ad-hoc spanning tree with the time obtained with other structures used for broadcasting in parallel and distributed systems. This comparison is done with an MPI emulation.

[1]  Dhabaleswar K. Panda,et al.  Communication modeling of heterogeneous networks of workstations for performance characterization of collective operations , 1999, Proceedings. Eighth Heterogeneous Computing Workshop (HCW'99).

[2]  Antony I. T. Rowstron,et al.  Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems , 2001, Middleware.

[3]  Bronis R. de Supinski,et al.  Exploiting hierarchy in parallel computer networks to optimize collective operation performance , 2000, Proceedings 14th International Parallel and Distributed Processing Symposium. IPDPS 2000.

[4]  David R. Karger,et al.  Chord: A scalable peer-to-peer lookup service for internet applications , 2001, SIGCOMM '01.

[5]  Bruce Lowekamp,et al.  ECO: Efficient Collective Operations for communication on heterogeneous networks , 1996, Proceedings of International Conference on Parallel Processing.

[6]  Silvia Figueira,et al.  Improving Binomial Trees for Broadcasting in Local Networks of Workstations 1 , 2002 .

[7]  Massimo Bernaschi,et al.  Collective communication operations: experimental results vs. theory , 1998, Concurr. Pract. Exp..

[8]  Scott Shenker,et al.  Routing Algorithms for DHTs: Some Open Questions , 2002, IPTPS.

[9]  Henri E. Bal,et al.  MagPIe: MPI's collective communication operations for clustered wide area systems , 1999, PPoPP '99.

[10]  Bruce S. Davie,et al.  Computer Networks: A Systems Approach , 1996 .

[11]  Massimo Bernaschi,et al.  Collective communication operations: experimental results vs. theory , 1998 .

[12]  Richard Wolski,et al.  Dynamically forecasting network performance using the Network Weather Service , 1998, Cluster Computing.

[13]  Henri E. Bal,et al.  Bandwidth-efficient collective communication for clustered wide area systems , 2000, Proceedings 14th International Parallel and Distributed Processing Symposium. IPDPS 2000.

[14]  Edith Cohen,et al.  Search and replication in unstructured peer-to-peer networks , 2002, ICS '02.

[15]  Richard Wolski,et al.  The network weather service: a distributed resource performance forecasting service for metacomputing , 1999, Future Gener. Comput. Syst..

[16]  Ian T. Foster,et al.  A Grid-Enabled MPI: Message Passing in Heterogeneous Distributed Computing Systems , 1998, Proceedings of the IEEE/ACM SC98 Conference.

[17]  Dennis Gannon,et al.  On the Impact of Communication Complexity on the Design of Parallel Numerical Algorithms , 1984, IEEE Transactions on Computers.

[18]  George K. Thiruvathukal,et al.  Wide-Area Implementation of the Message Passing Interface , 1998, Parallel Comput..

[19]  R. Sarnath,et al.  Proceedings of the International Conference on Parallel Processing , 1992 .