Experience with Building Reliable RPC Servers Based on a Fault Tolerant RPC Library

Abstract

This paper presents a fault tolerant RPC mechanism based on Sun RPC and IP multicast protocol. The fault tolerant RPC mechanism is provided as an RPC library, called libFIRPC, whose interface is compatible with that of conventional Sun's RPC library. Thanks to this compatibility, a reliable RPC server can be developed in the same way a conventional RPC server is constructed. The service reliability is ensured by replicating the server to a group of server replicas. Coordinator-cohort replication model in conjunction with read-one/write-all policy is used to guarantee state consistency between the server replicas. We also present our experience with building several reliable servers, including the portmapper and NFS server, over our fault tolerant RPC mechanism. The results of performance evaluation shows that the reliable NFS server has the same performance as the single unreplicated NFS server has when serving read-only requests.