2013/04/05 by Jian-Jia Chen, Chen, Jian-Jia, Mong-Jen Kao +7
Computer Science · #Data Structures and Algorithms (cs.DS) #FOS: Computer and information sciences #cs.DS
paper · pdf · doi:10.48550/arxiv.1304.1590
arxiv created 2013/04/08 · arxiv updated 2013/04/09
We consider the problem of online dynamic power management that provides hard real-time guarantees. In this problem, each of the given jobs is associated with an arrival time, a deadline, and an execution time, and the objective is to decide a schedule of the jobs as well as a sequence of state transitions on the processors so as to minimize the total energy consumption. In this paper, we examine the problem complexity and provide online strategies to achieve energy-efficiency. First, we show that the competitive factor of any online algorithm for this problem is at least 2.06. Then we present an online algorithm which gives a 4-competitive schedule. When the execution times of the jobs are unit, we show that the competitive factor improves to 3.59. At the end, the algorithm is generalized to allow a trade-off between the number of processors we use and the energy-efficiency of the resulting schedule.