Evaluation of interval-based dynamic voltage scaling algorithms on mobile Linux system

Authors:
Haklin Kimm;Sung Shin;Chang Oan Sung
Affiliations:
East Stroudsburg University of Penn, East Stroudsburg, PA;South Dakota State University, Brookings, SD;Indiana University Southeast, New Albany, IN
Venue:
Proceedings of the 2007 ACM symposium on Applied computing
Year:
2007

Citing 8
Cited 0

Comparing algorithm for dynamic speed-setting of a low-power CPU

MobiCom '95 Proceedings of the 1st annual international conference on Mobile computing and networking
Computer architecture (2nd ed.): a quantitative approach

Computer architecture (2nd ed.): a quantitative approach
The simulation and evaluation of dynamic voltage scaling algorithms

ISLPED '98 Proceedings of the 1998 international symposium on Low power electronics and design
Performance Comparison of Dynamic Voltage Scaling Algorithms for Hard Real-Time Systems

RTAS '02 Proceedings of the Eighth IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'02)
Automatic monitoring for interactive performance and power reduction

Automatic monitoring for interactive performance and power reduction
Application-directed voltage scaling

IEEE Transactions on Very Large Scale Integration (VLSI) Systems - Special section on low power
Scheduling for reduced CPU energy

OSDI '94 Proceedings of the 1st USENIX conference on Operating Systems Design and Implementation
SimDVS: an integrated simulation environment for performance evaluation of dynamic voltage scaling algorithms

PACS'02 Proceedings of the 2nd international conference on Power-aware computer systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

During the last several years, dynamic voltage scaling (DVS) algorithms are being used for energy consumption on real, fully functional battery supplied devices, adjusting the clock speed and supply voltage dynamically. Most DVS algorithms are investigated in interval-based and task-based strategies. Task-based algorithms consider task information, especially task deadline, on deciding what speed to choose at any given time. Interval-based algorithms predict the CPU speed of the upcoming interval based on observations of the CPU utilization of previous intervals, and then set the speed for that interval based on this prediction. Most DVS algorithms have only been tested in simulation environments. In this paper, those interval-based DVS algorithms are modified with different parameters on different workloads, and evaluated to know which one saves the most energy while not degrading computer performance.