Studies on striping and buffer caching issues for the software RAID file system

Abstract

A software RAID file system is defined as a system that distributes data redundantly across an array of disks attached to each of the workstations connected on a high-speed network. This configuration provides higher throughput and availability compared to conventional file systems. In this paper, we consider two specific issues regarding the distribution of data among the cluster, namely, striping and buffer caching for such an environment. Through simulation studies we compare the performance of various striping methods and show that for effective striping in software RAID file systems, it must take advantage of its flexible nature. Further, for buffer caching, we show that conventional caching schemes developed for distributed systems are insufficient, and that the Exclusively Old Data and Parity scheme that is presented in this paper, overcomes the limitations of the previously proposed schemes. (C) 2002 Elsevier Science B.V. All rights reserved.