Techniques for an Energy Aware Parallel File System

The decreasing cost of disk drives has made it possible to maintain large-scale storage systems and to achieve high performance for I/O demanding applications. One possible optimization that can be applied to these types of systems is to move the I/O load to a subset of the disks; so that the commonly accessed data resides in a set of 'popular' disks, whereas the 'unpopular' disks can stay in low energy modes. There have been several approaches to this kind of optimization both at the hardware and the software level. In this paper, we propose techniques for an Energy-Aware Parallel File System and compare different server assignment configurations to understand what kind of improvements can be achieved at the file system level to save energy while maintaining acceptable performance.