Performance Evaluation of Fault Tolerance for Parallel Applications in Networked Environments

Pierre Sens (1) , Bertil Folliot (1)

1 SRC - Systèmes répartis et coopératifs

This paper presents the performance evaluation of a software fault manager for distributed applications. Dubbed STAR, it uses the natural redundancy existing in networks of workstations to offer a high level of fault tolerance. Fault management is transparent to the supported parallel applications. STAR is application independent, highly configurable and easily portable to UNIX-like operating systems. The current implementation is based on independent checkpointing and message logging. Measurements show the efficiency and the limits of this implementation. The challenge is to show that a software approach to fault tolerance can efficiently be implemented in a standard networked environment.

Domaines: Informatique [cs]

Pierre Sens, Bertil Folliot. Performance Evaluation of Fault Tolerance for Parallel Applications in Networked Environments. 26th International Conference on Parallel Processing, Aug 1997, Bloomington, IL, United States. pp.334-341, ⟨10.1109/ICPP.1997.622663⟩. ⟨hal-01629551⟩