The CORBA Object Group Service (OGS) is a new CORBA service that provides support for fault-tolerance through the replication of CORBA objects. In this paper, we present several optimization techniques that are used to improve the performance of OGS. For each optimization, we analyse the impact on the throughput and the response time of OGS. The optimization techniques presented in this paper are quite generic and can be applied to many faulttolerant distributed algorithms based on consensus.