Disk array storage system reliability

Fault tolerance requirements for near term disk array storage systems are analyzed. The excellent reliability provided by RAID Level five data organization is seen to be insufficient for these systems. The authors consider various alternatives - improved MTBF and MTTR times as well as smaller reliability groups and increased numbers of check disks per group - to obtain the necessary improved reliability. Two data organization schemes based on maximum distance separable error correcting codes are introduced. Several figures of merit are calculated using a standard Markov failure and repair model for these organizations. Based on these results, the multiple check disk approach to improved reliability is an excellent option.