Adaptive Load Balancing for Lookups in Heterogeneous DHT

In DHT based P2P networks, the heterogeneity of node capacity and the skewness in query pattern could result in lookup load imbalance. Indegree adjustment is an effective, light-weighted and heterogeneity-aware approach to balance lookup traffic in local area, but not globally. In this paper we present a hybrid algorithm combining in degree adjustment and virtual server (VS) approaches to address both local and global lookup imbalance in DHT networks. We divide local nodes into zones, and select strong nodes to form a virtual server group (VSG). Hot zones can apply for virtual servers from VSG in order to distribute load in system wide. Nodes and VSs in the zones use indegree adjustment to maintain zone balance. We can therefore significantly improve the load balancing of traffic in these systems, and consequently their performance.Results from experimental evaluation demonstrate the effectiveness of our approach.

[1]  Scott Shenker,et al.  Making gnutella-like P2P systems scalable , 2003, SIGCOMM '03.

[2]  Peter G. Kropf,et al.  Dynamic Load Sharing in Peer-to-Peer Systems: When Some Peers Are More Equal than Others , 2007, IEEE Internet Computing.

[3]  Richard M. Karp,et al.  Load balancing in dynamic structured P2P systems , 2004, IEEE INFOCOM 2004.

[4]  Kam-Wing Ng,et al.  Achieving Resilient and Efficient Load Balancing in DHT-based P2P Systems , 2006, Proceedings. 2006 31st IEEE Conference on Local Computer Networks.

[5]  Cheng-Zhong Xu,et al.  Elastic Routing Table with Provable Performance for Congestion Control in DHT Networks , 2010, IEEE Trans. Parallel Distributed Syst..

[6]  Margo I. Seltzer,et al.  Distributed, secure load balancing with skew, heterogeneity and churn , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[7]  Cheng-Zhong Xu,et al.  Elastic Routing Table with Provable Performance for Congestion Control in DHT Networks , 2006, 26th IEEE International Conference on Distributed Computing Systems (ICDCS'06).

[8]  Peter Druschel,et al.  Pastry: Scalable, distributed object location and routing for large-scale peer-to- , 2001 .

[9]  David R. Karger,et al.  Simple Efficient Load Balancing Algorithms for Peer-to-Peer Systems , 2004, IPTPS.

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

[11]  Richard M. Karp,et al.  Load Balancing in Structured P2P Systems , 2003, IPTPS.

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

[13]  Yiming Hu,et al.  Efficient, proximity-aware load balancing for DHT-based P2P systems , 2005, IEEE Transactions on Parallel and Distributed Systems.