Improving Chord Network Performance Using Geographic Coordinates
Structured peer-to-peer overlay networks such as Chord, CAN, Tapestry, and Pastry, operate as distributed hash tables (DHTs). However, since every node is assigned a unique identifier in the basic design of DHT (randomly hashed), ”locality-awareness” is not inherent due to the topology mismatching between the P2P overlay network and the physical underlying network. In this paper, we propose to incorporate physical locality into a Chord system. To potentially benefit from some level of knowledge about the relative proximity between peers, a network positioning model is necessary for capturing physical location information of network nodes. Thus, we incorporate GNP (Global Network Positioning) into Chord (Chord-GNP) since peers can easily maintain geometric coordinates that characterize their locations in the Internet. Next, we identify and explore three factors affecting Chord-GNP performance: distance between peers, message timeout calculation and lookup latency. The measured results show that Chord-GNP efficiently locates the nearest available node providing a locality property. In addition, both the number of the messages necessary to maintain routing information and the time taken to retrieve data in Chord-GNP is less than that in Chord.
Unable to display preview. Download preview PDF.
- 1.Stoica, I., Morris, R., Karger, D., Kaashock, M., Balakrishman, H.: Chord: A scalable P2P lookup protocol for Internet applications. In: Proc. of ACM SIGCOMM (2001)Google Scholar
- 2.Ratnasamy, S., Francis, P., Handley, M., Karp, R., Shenker, S.: A scalable content addressable network. In: Proc. of ACM SIGCOMM (August 2001)Google Scholar
- 3.Rowstron, A., Druschel, P.: Pastry: Scalable, decentralized object location and routing for large-scale p2p systems. In: Proc. of IFIP/ACM Middleware (2001)Google Scholar
- 4.Zhao, B., Huang, L., Stribling, J., Rhea, S.C., Joseph, A., Kubiatowicz, J.: Tapestry: A global-scale overlay for rapid service deployment. IEEE J-SAC 22(1) (2004)Google Scholar
- 5.Castro, M., Druschel, P., Hu, Y.C., Rowstron, A.: Exploiting Network Proximity in Peer-to-Peer Overlay Networks. In: International Workshop on Future Directions in Distributed Computing (FuDiCo), Bertinoro, Italy (June 2002)Google Scholar
- 6.Hong, F., Li, M., Yu, J., Wang, Y.: PChord: Improvement on Chord to Achieve Better Routing Efficiency by Exploiting Proximity. In: ICDCS Workshops 2005, pp. 806–811 (2005)Google Scholar
- 7.Karger, D., Lehman, E., Leighton, T., Panigrahy, R., Levine, M., Lewin, D.: Consistent hashing and random trees: distributed caching protocols for relieving hot spots on the world wide web. In: Proceedings of the twenty-ninth annual ACM symposium on Theory of computing, May 1997, pp. 654–663 (1997)Google Scholar
- 8.Secure hash standard, NIST, U.S. Dept. of Commerce, National Technical Information Service FIPS 180-1 (April 1995)Google Scholar
- 9.Ng, T.S.E., Zhang, H.: Predicting internet network distance with coordinates-based approaches. In: Proceedings of IEEE Infocom (May 2002)Google Scholar
- 10.Ng, T.S.E., Zhang, H.: Towards Global Network Positioning. In: ACM SIGCOMM Internet Measurement Workshop, San Francisco, CA (November 2001)Google Scholar
- 11.Overlay Weaver: an overlay construction toolkit, http://overlayweaver.sf.net/
- 12.Shudo, K., Tanaka, Y., Sekiguchi, S.: Overlay Weaver: An overlay construction toolkit Computer Communications (Special Issue on Foundations of Peer-to-Peer Computing) 31(2), 402–412 (2008) (available online on August 14, 2007)Google Scholar