Checkpointing and rollback-recovery algorithms in distributed systems

Deng, Yong; Park, E. K.

doi:10.1016/0164-1212(94)90057-4

Cited by 24 publications

(38 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the above algorithms follow two approaches to reduce the overhead associated with coordinated checkpointing algorithms: One is to minimize the number of synchronization messages and the number of checkpoints [2], [5], [6], [9], [10], [13], [19], the other is to make checkpointing nonblocking [4], [7], [11], [20]. These two approaches were orthogonal in previous years until the Prakash-Singhal algorithm [18] combined them.…”

Section: Related Workmentioning

confidence: 99%

“…However, these two algorithms [10], [13] assume a complex scheme (such as slide window) to deal with the message loss problem and do not consider lost messages in checkpointing and recovery. Deng and Park [6] proposed an algorithm to address both orphan message and lost messages. In Koo and Toueg's algorithm [10], if any of the involved processes is not able to or not willing to take a checkpoint, the entire checkpointing process is aborted.…”

Section: Related Workmentioning

confidence: 99%

“…Most of the existing coordinated checkpointing algorithms [6], [10], [13] rely on the two-phase protocol and save two kinds of checkpoints on stable storage: tentative and permanent. In the first phase, the initiator takes a tentative checkpoint and forces all relevant processes to take tentative checkpoints.…”

Section: Basic Ideas Behind Nonblocking Algorithmsmentioning

confidence: 99%

“…All the above coordinated checkpointing algorithms [2], [5], [6], [9], [10], [13], [19] require processes to be blocked during checkpointing. Checkpointing includes the time to trace the dependence tree and to save the state of processes on stable storage, which may be long.…”

Section: Related Workmentioning

confidence: 99%

“…Much of the previous work [6], [9], [10], [13] in coordinated checkpointing has focused on minimizing the number of synchronization messages and the number of checkpoints during checkpointing. However, these algorithms (called blocking algorithms) force all relevant processes in the system to block their computations during checkpointing.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

On coordinated checkpointing in distributed systems

Singhal

1998

IEEE Trans. Parallel Distrib. Syst.

107

View full text Add to dashboard Cite

Abstract-Coordinated checkpointing simplifies failure recovery and eliminates domino effects in case of failures by preserving a consistent global checkpoint on stable storage. However, the approach suffers from high overhead associated with the checkpointing process. Two approaches are used to reduce the overhead: First is to minimize the number of synchronization messages and the number of checkpoints, the other is to make the checkpointing process nonblocking. These two approaches were orthogonal in previous years until the Prakash-Singhal algorithm [18] combined them. In other words, the Prakash-Singhal algorithm forces only a minimum number of processes to take checkpoints and it does not block the underlying computation. However, we found two problems in this algorithm. In this paper, we identify these problems and prove a more general result: There does not exist a nonblocking algorithm that forces only a minimum number of processes to take their checkpoints. Based on this general result, we propose an efficient algorithm that neither forces all processes to take checkpoints nor blocks the underlying computation during checkpointing. Also, we point out future research directions in designing coordinated checkpointing algorithms for distributed computing systems.

show abstract