Efficient caching scheme for software RAID file systems in workstation clusters

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 provides high throughput as well as higher availability. In this paper, we present an efficient caching scheme for the software RAID file system. The performance of this scheme is compared to two other schemes previously proposed for conventional file systems and adapted for the software RAID file system. As in hardware RAID systems, we found small-writes to be the performance bottleneck in software RAID file systems. To tackle this problem, we logically divided the cache into two levels. By keeping old data and parity values in the second-level cache we were able to eliminate much of the extra disk reads and writes necessary for write-back of dirty blocks. Using trace driven simulations we show that the proposed scheme improves performance by at least 10% for both the hit ratio and the average system busy time.