Controlling Energy Demand in Mobile Computing SystemsThis lecture provides an introduction to the problem of managing the energy demand of mobile devices. Reducing energy consumption, primarily with the goal of extending the lifetime of battery-powered devices, has emerged as a fundamental challenge in mobile computing and wireless communication. The focus of this lecture is on a systems approach where software techniques exploit state-of-the-art architectural features rather than relying only upon advances in lower-power circuitry or the slow improvements in battery technology to solve the problem. Fortunately, there are many opportunities to innovate on managing energy demand at the higher levels of a mobile system. Increasingly, device components offer low power modes that enable software to directly affect the energy consumption of the system. The challenge is to design resource management policies to effectively use these capabilities. |
Contents
1 | |
Management of Device Power States | 23 |
Dynamic Voltage Scheduling DVS | 39 |
Multiple DevicesInteractions and Tradeoffs | 55 |
EnergyAware Application Code | 69 |
Challenges and Opportunities | 79 |
Author Biography | 89 |
Other editions - View all
Controlling Energy Demand in Mobile Computing Systems Carla Schlatter Ellis No preview available - 2007 |
Common terms and phrases
access patterns ACPI active adapt algorithm battery lifetime Bluetooth breakeven buffer cache Chapter component control flow graph currentcy cycles deadlines demand dynamic slack dynamic voltage dynamic voltage scaling energy consumption energy management energy savings example execution exploit fidelity FIGURE file system frequency and voltage goal hardware hints idle gap idle periods Intel Corporation Intel XScale interactions interface interval laptop latency levels lower power memory metrics mobile computing mobile device monitoring multimeter multiple devices nodes operating system optimal performance Phigh platform policies power consumption power management power state transitions prediction prefetching problem processes processor radio RDRAM real-time Reprinted by permission resource simulation smartphone solution speed schedule spinning spinup storage switch system call system model task set task's techniques threshold tier timeout tradeoff trigger Turducken voltage scaling WCEC WCET wireless workload worst-case execution
Popular passages
Page 88 - W. Ye, N. Vijaykrishnan, M. Kandemir, and MJ Irwin. The design and use of simplepower: a cycle-accurate energy estimation tool.
Page 83 - Brooks D., Tiwari V., Martonosi M., "Wattch: a framework for architectural-level power analysis and optimizations", in: Proceedings of the 27th International Symposium on Computer Architecture (ISCA).
Page 84 - J. Flinn and M. Satyanarayanan. PowerScope: A tool for profiling the energy usage of mobile applications.
Page 87 - Shin, KG Real-time dynamic voltage scaling for low-power embedded operating systems, In Proceedings of ACM Symposium on Operating Systems Principles, December 2001, pp.
Page 85 - A Predictive System Shutdown Method for Energy Saving of Event-Driven Computation", in Proceedings of the Int.
Page 83 - S. Chandra and A. Vahdat. Application-specific Network Management for Energyaware Streaming of Popular Multimedia Formats.
Page 87 - McAuley. Energy is Just Another Resource: Energy Accounting and Energy Pricing in the Nemesis OS.
Page 85 - Online strategies for dynamic power management in systems with multiple power-saving states,
Page 86 - Non-ideal battery and main memory effects on CPU speed-setting for low power, IEEE Trans.
Page 85 - Using complete machine simulation for software power estimation: the SoftWatt approach", in: Proceedings of the International Symposium on High Performance Computer Architecture (HPCA-8), February 2002.