Design and Analysis of Capacity Extendmle Disk Array System: The Diagonal Move Algorithm

Abstract

With increased I/O performance and at least one disk failure tolerance, data redundant disk array as secondary storage system efficiently translate from a conventional computer storage system to be with better I/O rate, higher data transfer rate, and stronger reliability than traditional large single-disk systems. The increased I/O performance in measurement of I/O operating rate and data transfer rate are mostly gained from simultaneous data retrieval from several disks which are organized in parallel as shown in Figure 2-1. This parallel disks organization lets us have better I/O operating rate than the singular disk architecture, since it has several disk- I/O operations running concurrently. The I/O operating rate is defined as the number of I/O operations per second. And the simultaneous data access from disks lets us have better data transfer rate than any singular disk architecture, since we are retrieving data from more than one disk at the same time. The data transfer rate is defined as amount of data transferred through the bus or network per second such as bits per second (BPS). The disk data access time is the composition of seek time, rotation time, and data transfer time. Because data transfer speed over the bus or network electronically is much faster than the slow mechanical disk drives, accessing several disks in parallel can contribute to the bus or network having better utilization.