摘要： 高可靠性是计算机系统所要解决和实现的重要目标,而容错技术是提高可靠性的有效途径。文章论述了可靠性的基本概念和3种故 障模式;介绍了计算机容错设计所采用的硬件冗余、信息冗余和时间冗余技术,着重分析了RAID技术和纠错码技术;指出了计算机容错设 计的发展趋势及面临的挑战。

关键词: 可靠性, 容错, 冗余, 平均故障间隔时间(MTBF)

Abstract: High reliability is one of the main goals to be solved and realized for computer systems,and fault tolerance techniques are proposed as the efficient approach to attain high reliability. This paper discusses the basic concept of reliability and three kinds of fault modes. Then the redundancy techniques used for designing fault-tolerant computers are presented, which include hardware redundancy, information redundancy and time redundancy. Especially, the RAID and error correcting code technology are analyzed in detail. Finally, this paper points out the future development and present challenge of fault tolerance design.

中图分类号: 

• TP302