Event-driven power management

Abstract

Energy consumption of electronic devices has become a serious concern in recent years. Power management (PM) algorithms aim at reducing energy consumption at the system-level by selectively placing components into low-power states. Formerly, two classes of heuristic algorithms have been proposed for power management: timeout and predictive. Later, a category of algorithms based on stochastic control was proposed for power management. These algorithms guarantee optimal results as long as the system that is power managed can be modeled well with exponential distributions. We show that there is a large mismatch between measurements and simulation results if the exponential distribution is used to model all user request arrivals. We develop two new approaches that better model system behavior for general user request distributions. Our approaches are event driven and give optimal results verified by measurements. The first approach we present is based on renewal theory. This model assumes that the decision to transition to low power state can be made in only one state. Another method we developed is based on the Time-Indexed Semi-Markov Decision Process model (TISMDP). This model has wider applicability because it assumes that a decision to transition into a lower-power state can be made upon each event occurrence from any number of states. This model allows for transitions into low power states from any state, but it is also more complex than our other approach. It is important to note that the results obtained by renewal model are guaranteed to match results obtained by TISMDP model, as both approaches give globally optimal solutions. We implemented our power management algorithms on two different classes of devices: two different hard disks and client-server WLAN systems such as the SmartBadge [19] or a laptop. The measurement results show power savings ranging from a factor of 1¡ 7 up to 5¡ 0 with insignificant variation in performance.

A MAC layer power management scheme for efficient energy delay tradeoff in a WLAN

“…In addition, policies can be stationary (the same policy applies at any point in time) or non stationary (the policy changes over time). All stochastic approaches except for the discrete adaptive approach presented in [22] are stationary. Approaches based on discrete time setting require policy evaluation even in the low power state [21], [22]thus wasting energy.…”

“…All stochastic approaches except for the discrete adaptive approach presented in [22] are stationary. Approaches based on discrete time setting require policy evaluation even in the low power state [21], [22]thus wasting energy. On the other hand, event driven models based on exponential distributions [25] show little or no power savings when implemented in real systems since the exponential model does not describe well the request interarrival times [23], [24].…”

“…Stochastic models use distributions to describe the times between arrivals of user requests (inter-arrival times), the length of time it takes for a device to service a user's request and the time it takes for a device to transition between its power states. The system model for stochastic optimization can be described either with just memoryless distributions (exponential or geometric) [21], [22] or with general distributions [23], [24]. Power management policies can also be classified into two categories by the manner in which decisions are made: discrete time [15], [16] or event driven [23], [24].…”

Flow-Through-Queue based Power Management for Gigabit Ethernet Controller

“…A general distribution is assumed for the inter-arrival times because an exponential distribution would underestimate the occurrence probability for long request inter-arrival times and so it does not adequately model the request arrival time in the idle periods [14]. The service time behavior is captured by a given service time distribution for the functional module when it is in the active mode.…”

Transmittance Scaling for Reducing Power Dissipation of a Backlit TFT-LCD

“…Previous studies on battery-powered electronics point out that the display subsystem dominates the energy consumption of the whole system. In the SmartBadge system, for instance, the display consumes 29%, 29%, and 50% of the total power in the active, idle, and standby modes, respectively [1]. Direct-view LCDs can largely be categorized into reflective and transmissive displays which utilize ambient light and light from an artificial light source (e.g., fluorescent backlight tube) respectively.…”