Title

Multicomputer Checkpointing

Authors

Ken Wong, Washington University in St Louis
Mark Franklin, Washington University in St LouisFollow

Document Type

Technical Report

Publication Date

1992

Filename

WUCS-92-07.pdf

DOI:

10.7936/K7Q52MZ7

Technical Report Number

WUCS-92-07

Abstract

This paper examines the performance of synchronous checkpointing in a distributed computing environment with and without load redistribution. Performance models are developed, and optimum checkpoint intervals are determined. We extend earlier work by allowing for multiple nodes, state dependent checkpoint intervals, and a performance metric which is coupled with failure-free performance. We show that the optimum checkpoint intervals in the presence of load redestribution has a numerical solution in all cases and a closed form in many reasonable cases. These new results are then used to determine when performance can benefit load redistribution.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7Q52MZ7

Recommended Citation

Wong, Ken and Franklin, Mark, "Multicomputer Checkpointing" Report Number: WUCS-92-07 (1992). All Computer Science and Engineering Research.
https://openscholarship.wustl.edu/cse_research/519