Empirical analysis of power management schemes for multi-core smartphones

Authors:
Sangwook Kim;Hwanju Kim;Jongwon Kim;Joonwon Lee;Euiseong Seo
Affiliations:
Sungkyunkwan University, Suwon, South Korea;Korea Advanced Institute of Science and Technology, Daejeon, South Korea;Samsung Electronics Co., Suwon, South Korea;Sungkyunkwan University, Suwon, South Korea;Sungkyunkwan University, Suwon, South Korea
Venue:
Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication
Year:
2013

Citing 9
Cited 0

Real-time dynamic voltage scaling for low-power embedded operating systems

SOSP '01 Proceedings of the eighteenth ACM symposium on Operating systems principles
Power: A First-Class Architectural Design Constraint

Computer
Dynamic Power Optimization Targeting User Delays in Interactive Systems

IEEE Transactions on Mobile Computing
Operating System Modifications for Task-Based Speed and Voltage

Proceedings of the 1st international conference on Mobile systems, applications and services
User-Driven Frequency Scaling

IEEE Computer Architecture Letters
TSB: A DVS algorithm with quick response for general purpose operating systems

Journal of Systems Architecture: the EUROMICRO Journal
An analysis of power consumption in a smartphone

USENIXATC'10 Proceedings of the 2010 USENIX conference on USENIX annual technical conference
Dynamic voltage and frequency scaling: the laws of diminishing returns

HotPower'10 Proceedings of the 2010 international conference on Power aware computing and systems
Chameleon: operating system support for dynamic processors

ASPLOS XVII Proceedings of the seventeenth international conference on Architectural Support for Programming Languages and Operating Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Dynamic power management schemes in mobile devices such as smartphones and tablet PCs enhance battery life at the cost of prolonged user-perceived response time while the response time is a crucial factor for user experience. This paper presents systematic analysis of existing power management schemes adopted in recent smartphones in terms of user-perceived response time and energy consumption. For this analysis, we developed a latency measurement benchmark tool to quantify responsiveness to user inputs and used it with an externally-connected power meter to concurrently measure energy consumption and response latency. The evaluation showed that some existing DVFS schemes can significantly harm the response time. More seriously, the analysis revealed that the processor hotplug technique for multi-core systems may reduce responsiveness even without any gain in energy savings.