An evaluation of redundant arrays of disks using an Amdahl 5890

Abstract. Recently we presented several disk array architectures designed to increase the data rate and I/O rate of supercomputing applications, transaction processing, and file systems [Patterson 88]. In this paper we present a hardware performance measurement of two of these architectures, mirroring and rotated parity. We see how throughput for these two architectures is affected by response time requirements, request sizes, and read to write ratios. We find that for applications with large accesses, such as many supercomputing applications, a rotated parity disk array far outperforms traditional mirroring architecture. For applications dominated by small accesses, such as transaction processing, mirroring architectures have higher performance per disk than rotated parity architectures. 1. The I/O Crisis Over the past decade, processing speed, memory speed, memory capacity, and disk capacity have all grown tremendously: Single chip processors have increased in speed at the rate of 40%-100 % per year [Bell 84, Joy 85]. Caches have increased in speed 40 % to 100 % per year. Main memory has quadrupled in capacity every two o