关键词: 向前恢复 容错计算 故障恢复模型和策略 系统性能

Abstract: The system architecture of a kind of scalable fault-tolerant computer with triple module redundancy is pro- posed. Detailed study on its fault recovery model and scheme corresponding to different fault occurrence situations is done. Through synthetically

Key words: Forward recovery,Fault-tolerant computing, Fault recovery model and scheme,System performance