A distributed consistent global checkpoint algorithm with a minimum number of checkpoints
Yoshifumi Manabe · 2002
A distributed coordinated checkpointing algorithm is shown. A consistent global checkpoint is a set of states in which no message is recorded as received in one process and as not yet sent in another process. This algorithm obtains a consistent global checkpoint for any checkpoint initiation by any process. Under Chandy and Lamport's assumption that one consistent global checkpoint is obtained for a set of concurrent checkpoint initiations, the total number of checkpoints is minimized. This paper then modifies the assumption in order to reduce the number of checkpoints further.