Jacob R. Lorch and Alan Jay Smith
This paper describes RightSpeed, a task-based speed and voltage scheduler for Windows 2000. It takes advantage of the ability of certain processors, such as those from Transmeta and AMD, to dynamically change speed and voltage and thus to save energy while running more slowly. RightSpeed uses PACE, an algorithm that computes the most energy efficient way to meet task deadlines with high probability. Since most applications do not provide enough data about tasks, such as task deadlines, for PACE to work, RightSpeed uses simple and efficient heuristics to automatically detect task characteristics for such applications. RightSpeed has only 1.2% background overhead and its operations take only a few microseconds each. It even performs PACE calculation, which is quite complicated, in only 4.4 us on average due to our extensive optimizations. RightSpeed is effective at meeting performance targets set by applications to within 1.5%. Although the PACE calculator does not save energy for the current generation of processors due to their limited range of worthwhile speed and voltage settings, we expect future processors to have greater such ranges, enabling PACE to reduce CPU energy consumption by 6.1-8.7% relative to the best standard algorithm. Furthermore, with PACE, giving a processor the ability to run at additional, higher speeds and voltages reduces overall energy consumption.
In Proceedings of the First International Conference on Mobile Systems, Applications, and Services (MobiSys)
Publisher Association for Computing Machinery, Inc.
Copyright © 2007 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept, ACM Inc., fax +1 (212) 869-0481, or firstname.lastname@example.org. The definitive version of this paper can be found at ACM’s Digital Library --http://www.acm.org/dl/.
|Address||San Francisco, CA|