ISBN 978-1-4577-1489-4International audienceSelf-convergence is a property of distributed systems, allowing a system, when it was perturbed or badly initialised, to recover a correct operation in a finite number of calculation steps. In this paper is explored the intrinsic robustness of a self converging algorithm with respect to soft errors resulting from SEU (Single Event Upset) phenomena. This study was performed by fault injection using a devoted test platform. A self-converging benchmark program was executed by a LEON3 processor implemented in an FPGA. The low number of observed errors puts in evidence the fault tolerance of the tested algorithm
International audienceThe single-event upset (SEU) fault tolerance of a benchmark self-converging algorithm is evaluated by fault injection campaigns performed using a devoted test platform. The number of observed errors significantly decreases depending on adopted implementation strategies
The robustness with respect to SEUs (SingleEvent Upset) of a self-converging algorithm is improved by fault-tolerance techniques implemented at software level. SEU-sensitivity evaluation was done by fault injection campaigns performed using a devoted test platform. Experimental results show that implementing faulttolerance by modifying the assembly code leads to significant improvements of the fault tolerance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.