Implementing Precise Interrupts in Pipelined Processors
IEEE Transactions on Computers
Error Recovery in Shared Memory Multiprocessors Using Private Caches
IEEE Transactions on Parallel and Distributed Systems
A low-overhead coherence solution for multiprocessors with private cache memories
ISCA '84 Proceedings of the 11th annual international symposium on Computer architecture
Hi-index | 0.00 |
This paper examines the problem of recovering from processor transient faults in pipelined multiprocessor systems. A pipelined machine allows out of order instruction execution and branch prediction to increase performance, thus a precise computation state may not be available. We propose a modified scheme to implement the precise computation state in a pipelined machine. The goal of this research is to implement checkpointing and rollback for error recovery in a pipelined system based on the technique to achieving precise computation state. Detailed analysis has been performed to demonstrate the effectiveness of this method.