Abstract Performance Analysis of Disk Arrays Under Failure*

Disk arrays (RAID) have been proposed as a possi-ble approach to solving the emerging I/O bottleneck problem. The performance of a RAID system when all disks are operational and the MTTF,,, (mean time to system failure) have been well studied. How-ever, the performance of disk arrays in the presence of failed disks has not received much attention. The same techniques that provide the storage efficient re-dundancy of a RAID system can also result in a sig-nificant performance hit when a single disk fails. This is of importance since single disk failures are expected to be relatively frequent in a system with a large num-ber of disks. In this paper we propose a new varia-tion of the RAID organization that has significant ad-vantages in both reducing the magnitude of the per