Fault Tolerance Techniques in Distributed System

Abstract — Fault tolerance is an important issue in distributed computing. Developers of early distributed systems took a simplistic approach to providing fault tolerance: They just used another copy of the same hardware as a backup. There are various factors & critical issues responsible for these overheads. This paper provides a study of fault tolerance techniques in distributed systems, especially replication and checkpointing. We also suggested fault tolerance by combining replication and checkpointing and implemented it in Java RMI. This work will provide a good reference for researchers.