I/O systems are increasingly becoming a major performance limitation to faster computer systems. 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. 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, request size, and the ratio of reads and writes. We also explore tradeoffs in the unit of interleaving and number of disks. We find that for applications with large accesses, such as many supercomputing applications, a rotated parity disk array far outperforms traditional mirroring architecture. In contrast, for applications with many small accesses, such as transaction processing and traditional file systems, mirroring disk arrays outperform rotated parity disk arrays.