IPDS '98 Final Version; Coordinated Checkpointing Without Direct Coordination

Coordinated checkpointing is a well-known method to achieve fault tolerance in distributed systems. Long running parallel applications and high-availability applications are two potential users of checkpointing, although with different requirements. Parallel applications need low failure-free overheads, and high-availability applications require fast and bounded recoveries. In this paper, we describe a new coordinated checkpoint protocol capable of satisfying both types of applications. The protocol uses time to avoid all types of direct coordination (e.g., message exchanges and message tagging), reducing the overheads to almost a min-imum. To ensure that rapid recoveries can be attained, the protocol guarantees small checkpoint latencies. The pro-tocol was implemented and tested on a cluster of worksta-tions connected by a 155 Mbit/sec ATM. Experimental re-sults show that the protocol overheads are very small. I