Optimistic Replication for Internet Data Services

We present a new replication algorithm that supports replication of a large number of objects on a diverse set of nodes. The algorithm allows replica sets to be changed dynamically on a per-object basis. It tolerates most types of failures, including multiple node failures, network partitions, and sudden node retirements. These advantages make the algorithm particularly attractive in large cluster-based data services that experience frequent failures and configuration changes. We prove the correctness of the algorithm and show that its performance is near-optimal. This report originally appeared in the 14th International Conference in Distributed Computing (DISC), October 2000, Toledo, Spain. The proceedings of the conference are published as Lecture Notes in Computer Science No. 1914, Springer Verlag.