Abstract
Checkpointing reduces loss of computation in the presence of failures. Two metrics characterize a checkpointing scheme: checkpoint overhead and checkpoint latency. This paper shows that a large increase in latency is acceptable if it is accompanied by a relatively small reduction in overhead. Also, for equidistant checkpoints, optimal checkpoint interval is shown to be typically independent of checkpoint latency.
Original language | English (US) |
---|---|
Pages (from-to) | 942-947 |
Number of pages | 6 |
Journal | IEEE Transactions on Computers |
Volume | 46 |
Issue number | 8 |
DOIs | |
State | Published - 1997 |
Externally published | Yes |
Keywords
- Checkpoint latency
- Checkpoint overhead
- Checkpointing and rollback
- Performance analysis
ASJC Scopus subject areas
- Software
- Theoretical Computer Science
- Hardware and Architecture
- Computational Theory and Mathematics