User Level SCSI Disk Feature Extraction

Despite the increases in the disk capacity and decreases in mechanical delays in recent years, the performance gap between magnetic disks and CPU continues to increase. To effectively improve disk performance, operating systems and file systems must have detailed information (e.g., zoning, bad-sector positions, and cache size) about the disks that they use. In this paper, we present a tool that we have developed to extract such information. We first specify the disk features that we extract. We then explain various interrogative and empirical methods for extracting these disk features. Finally, we present extraction results for two testbeds. From our empirical study, we conclude that intelligent data placement and access methods can be devised, by exploiting low-level disk knowledge, to attain better disk performance.