Modeling and Performance of MEMS-Based Storage Devices

MEMS-based storage devices are seen by many as promis-ing alternatives to disk drives. Fabricated using conven-tional CMOS processes, MEMS-based storage consists of thousands of small, mechanical probe tips that access gi-gabytes of high-density, nonvolatile magnetic storage. This paper takes a rst step towards understanding the perfor-mance characteristics of these devices by mapping them onto a disk-like metaphor. Using simulation models based on the mechanics equations governing the devices ' operation, this work explores how dierent physical characteristics (e.g., actuator forces and per-tip data rates) impact the design trade-os and performance of MEMS-based storage. Overall results indicate that average access times for MEMS-based storage are 6.5 times faster than for a modern disk (1.5 ms vs. 9.7 ms). Results from lesystem and database bench-marks show that this improvement reduces application I/O stall times up to 70%, resulting in overall performance im-provements of 3X. 1