Reliability of replicated data objects
- 7 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 402-406
- https://doi.org/10.1109/pccc.1989.37422
Abstract
When fault tolerance is improved by replicating data at several sites, a protocol is required to maintain the consistency of the data object in the presence of site failures. The most commonly used scheme is voting, which is unaffected by network partitions. When network partitions cannot occur, better performance can be achieved with available copy protocols. The authors investigate the reliability of replicated data objects managed by voting, available copy, and their variants. Where possible, closed-form expressions for the reliability of the various consistency protocols are derived using standard Markovian assumptions. In other cases, numerical solutions are found and validated with simulation results.Keywords
This publication has 6 references indexed in Scilit:
- Integrating static and dynamic voting protocols to enhance file availabilityPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Replication and fault-tolerance in the ISIS systemPublished by Association for Computing Machinery (ACM) ,1985
- A recovery algorithm for a distributed database systemPublished by Association for Computing Machinery (ACM) ,1983
- Weighted voting for replicated dataPublished by Association for Computing Machinery (ACM) ,1979
- Reliability modeling techniques for self-repairing computer systemsPublished by Association for Computing Machinery (ACM) ,1969
- Approximation Formulas for Reliability with RepairIEEE Transactions on Reliability, 1963