Abstract
This paper proposes a distributed hybrid fault tolerant model for mobile grid systems. It consists of proactive and reactive fault tolerant approaches. Before resource allocation, all the jobs are sorted and assigned to different level based on precedence. The reliability and failure probability of each resource is estimated. In case of proactive approach, the high priority jobs are assigned to the resources with high reliabilities. In reactive approach, a backup node with low failure probability will have the replica of jobs assigned for the resources with high failure probability. If a failure is detected in the primary resource, the backup resource will perform failure recovery by executing the remaining job.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call