Exploiting the Spring Structure of MEMS-Based Storage Devices to Reduce Their Shutdown Energy

New storage devices based on Micro-Electro-Mechanical Systems (MEMS) have been proposed [3]. Enabled by high storage densities (> 1 Tb/in), MEMS technology promises to deliver small-form factor, high-capacity, and low-power storage devices. These devices have potentially low cost, because they can be manufactured using batch MEMS fabrication technology. A MEMS-based storage device dissipates an order of magnitude less power than a disk drive. However, like disk drives, MEMS devices have a moving medium. For optimal energy saving, a MEMS device should be shut down (i.e., the sled is stopped) during periods of inactivity. Because of their micro-mechanical nature, MEMS devices lend themselves to more aggressive shut down policies than disk drives. As the timeout decreases, the number of shutdowns increases, because more periods of inactivity are exploited. As a result, the energy consumed to shut down increases and so the total energy, which limits the applicability of aggressive shutdown decisions. Interestingly, as Figure 1 shows, in MEMS devices the moving medium is suspended by springs. We show that the potential energy stored in the springs can be exploited at shutdown to accelerate the sled toward the center as shown in Figure 2. External energy is invested only to decelerate the sled so that it stops at the center. Consequently, the shutdown energy is reduced to the deceleration energy, allowing to increase the aggressiveness of the shutdown decisions. This energy benefit, however, comes at a performance cost; that is the sled takes long time to reach the center, since it is not actively accelerated. In this work, we compare the energy-efficient (EE) policy against a performance-efficient (PE) policy. The PE policy uses the actuators to accelerate for some distance as shown in Figure 2 and then decelerate the sled so that it reaches the center in the shortest time possible. When deploying the PE policy, the sled acts as if it is seeking from its current position to the center with the exception that it stops along Y as well as along X at the center. Therefore, energy is consumed during both acceleration and deceleration. We propose an analytical model of the EE policy, comparing it to the PE policy. We use state-of-the-art parameter settings of the IBM MEMS device.