An index-based checkpointing algorithm for autonomous distributedsystems

摘要

The paper presents an index based checkpointing algorithm fordistributed systems with the aim of reducing the total number ofcheckpoints while ensuring that each checkpoint belongs to at least oneconsistent global checkpoint (or recovery line). The algorithm is basedon an equivalence relation defined between pairs of successivecheckpoints of a process which allows, in some cases, to advance therecovery line of the computation without forcing check points in otherprocesses. This protocol shows good performance, especially inautonomous environments, where each process does not have any privateinformation about other processes