Experimental Evaluation of the Fault Tolerance of an Atomic Multicast Protocol

J. Arlat, M. Aguera, Y. Crouzet, J.-C. Fabre, E. Martins and D. Powell

 

Abstract

This paper presents a study that contributes to the validation of a dependable local area network proving multipoint communication services based on an atomic multicast protocol. This protocol is implemented in specialized communication servers that exhibit the fail-silent property, i.e., a kind of halt-on-failure behavior enforced by self-checking hardware. The tests that have been carried out utilize physical fault injection and have two objectives: i) to estimate the coverage of the self-checking mechanisms of the communication servers and ii) to test the properties that characterize the service provided by the aotmic multicast protocol in the presence of faults. An appreciable part of the paper is devoted to the description of the testbed that has been developed to carry out the fault injection experiments. The major results are presented and analyzed.