Scalable replication in database clusters

Abstract

In this paper, we explore data replication protocols that provide both fault tolerance and good performance without compromising consistency. We do this by combining transactional concurrency control with group communication primitives. In our approach, transactions are executed at only one site so that not all nodes incur in the overhead of producing results. To further reduce latency, we use an optimistic multicast technique that overlaps transaction execution with total order message delivery. The protocols we present in the paper provide correct executions while minimizing overhead and providing higher scalability.