ACHIEVING FAULT-TOLERANCE IN PIPELINED MULTIPROCESSOR SYSTEMS

This paper focuses on recovering from processor transient faults in pi pelined multiprocessor systems. A pipelined machine may employ out of order execution and branch prediction techniques to increase performan ce, thus a precise computation state would not be available. We propos e an efficient scheme to maintain the precise computation state in a p ipelined machine. The goal of this paper is to implement checkpointing and rollback recovery utilizing the technique of precise interrupt in a pipelined system. Detailed analysis is included to demonstrate the effectiveness of this method.