Replica Management Should Be A Game


	The OceanStore Project Paper Abstract Introspective Failure Analysis: Avoiding Correlated Failures in Peer-to-Peer Systems Hakim Weatherspoon, Tal Moscovitz, and John Kubiatowicz. Appears in Proceedings of International Workshop on Reliable Peer-to-Peer Distributed Systems 2002,

Failure independence is an important assumption for many fault tolerance techniques. Unfortunately, real system exhibit correlated failures. In this paper, we present a framework for online discovery of groups of server nodes that are maximally independent in their failure characteristics. We discuss the framework in detail and provide a preliminary evaluation.

[ Paper: PDF (88K), Postscript (165K), Compressed Postscript (53K) ]

Last modified on 07/29/2002 by Hakim Weatherspoon.