Low-power task scheduling for multiple devices

Abstract

Power management saves power by shutting down idle devices. These devices often serve requests from concurrently running tasks. Ordering task execution can adjust the lengths of idle periods and exploit better opportunities for power management. This paper presents an on-line low-power scheduling algorithm for multiple devices. Simulations show that it can save up to 33% power and reduce 40% state-transition delays. This algorithm is robust under imperfect knowledge of future requests and timing constraints; therefore, it is applicable to interactive systems.

Task Merging for Dynamic Power Management of Cyclic Applications in Real-Time Multiprocessor Systems

“…Among them, the authors of [10] proposed a scheduling algorithm with the consideration of min/max timing and min/max power constraints. In reference [11], an on-line scheduling algorithm is proposed to reduce the number of power mode switching in a multiprocessor system. However, precedence and deadline constraints are not considered during the optimization.…”

Power-aware scheduling and dynamic voltage setting for tasks running on a hard real-time system

“…The optimization problem is to minimize E sys with respect to constraints (2-1) to (2)(3)(4)(5)(6)(7)(8). Note that in this formulation, we ignore the energy and timing overhead associated with the voltage changes because switching of the CPU voltage normally takes between 10-100 microseconds depending on the hardware support for the DVS function.…”

“…The first one simply ignores the energy components introduced by idlene k in equation (2)(3)(4)(5)(6)(7)(8)(9). The optimization problem thus becomes a linear programming problem over continuous variable x(u,i), which can be solved in polynomial time.…”

“…The second approach introduces new 0-1 integer variables W k (v,h) to approximate the idle time it k,v as follows: And equation (2)(3)(4)(5)(6)(7)(8)(9) becomes an integer linear cost function. As a result, the original optimization problem is approximated with a mixed integer linear program.…”

“…An online scheduling algorithm for independent tasks is presented in [5]. This algorithm attempts to reduce the number of device on/off transitions by greedily extending the pattern for current device usage so as to reduce average power consumption in the near future.…”