A FAULT-TOLERANT MECHANISM FOR DISTRIBUTED/PARALLEL SYSTEM BASED ON TASK REPLICATION TECHNIQUES

摘要

In this work we propose a fault-tolerant mechanism for parallel programs based on task replication. We use a sequential discrete- event simulator of a distributed system subject to failures to compare our semi-active and passive approaches. The performance has been studied as a function of the program size, the processor degradation, and the system size. In addition, we compare our fault-tolerant mechanism with similar approaches. In this case, the criteria studied are the recovery time, the throughput, and the response time.