On Checkpoint Overhead in Distributed Systems Providing Session Guarantees

This paper presents the performance evaluation of the checkpointing and rollback-recovery protocols for distributed mobile systems, guarantying client-centric consistency models, despite failures of servers. The performance of considered protocols is evaluated by estimating the checkpointing overhead. Additionally, the paper disscusses the influence of the consistency models provided by the system on the moments of taking checkpoints. The impact of the obtained results on strategies for checkpointing in the mobile environment guarantying client-centric consistency models is assessed.

[1]  Kai Li,et al.  Diskless Checkpointing , 1998, IEEE Trans. Parallel Distributed Syst..

[2]  Gerard Tel,et al.  Introduction to Distributed Algorithms: Contents , 2000 .

[3]  Xiaola Lin,et al.  A Variational Calculus Approach to Optimal Checkpoint Placement , 2001, IEEE Trans. Computers.

[4]  Marvin Theimer,et al.  Session guarantees for weakly consistent replicated data , 1994, Proceedings of 3rd International Conference on Parallel and Distributed Information Systems.

[5]  Edward G. Coffman,et al.  Optimal fault-tolerant computing on multiprocessor systems , 1997, Acta Informatica.