Replication Control in Distributed File Systems

We present a replication control protocol for distributed file systems that can guarantee strict consistency or sequential consistency while imposing no performance overhead for normal reads. The protocol uses a primary-copy scheme with server redirection when concurrent writes occur. It tolerates any number of component omission and performance failures, even when these lead to network partition. Failure detection and recovery are driven by client accesses. No heartbeat messages or expensive group communication services are required. We have implemented the protocol in NFSv4, the emerging Internet standard for distributed filing.