Epidemic quorums for managing replicated data

Abstract

In the epidemic model an update is initiated on a single site and is propagated to other sites in a lazy manner. When combined with version vectors and event logs, this propagation mechanism delivers updates in causal order despite communication failures. We integrate quorums into the epidemic model to process transactions on replicated data while ensuring global serializability. We present a detailed simulation of a distributed replicated database and demonstrate the performance improvements.