1 / 20

PowerScope: A Tool for Profiling the Energy Usage of Mobile Applications

PowerScope: A Tool for Profiling the Energy Usage of Mobile Applications. Jason Flinn M. Satyanarayanan Carnegie Mellon University. Motivation. Energy is a critical resource for mobile computing. Comprehensive approach to energy management should include OS and apps , not just hardware!

metea
Download Presentation

PowerScope: A Tool for Profiling the Energy Usage of Mobile Applications

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. PowerScope: A Tool for Profiling the Energy Usage of Mobile Applications Jason Flinn M. Satyanarayanan Carnegie Mellon University

  2. Motivation • Energy is a critical resource for mobile computing. • Comprehensive approach to energy management • should include OS and apps, not just hardware! • PowerScope profiles the energy usage of applications. • Exposes targets for optimizations. • Quantifies benefits of improvements. • Iterative development reduces energy usage.

  3. PowerScope Design Overview • Profile generation is a two-stage process. • First stage: sample collection • Digital multimeter samples power levels. • Kernel instrumentation samples system activity. • Second stage: off-line analysis • Symbol tables and samples used to generate energy profile.

  4. Stage 1: Sample Collection System Monitor samples system activity. Digital multimeter samples current levels. Energy Monitor stores current samples. Hardware interface correlates sampling.

  5. Stage 2 Off-Line Analysis Generates profile of energy usage. Runs off-line to reduce overhead Attributes samples to specific processes and functions.

  6. Sample Profile: Summary Elapsed Total Average Process Time (s)Energy (J)Power (W) ------------------------------ ---------------------------------------- /usr/odyssey/bin/xanim 66.57 643.179.66 /usr/X11R6/bin/X 35.72 331.589.28 /netbsd (kernel) 50.89 328.716.46 Interrupts-WaveLAN 18.62 165.88 8.91 /usr/odyssey/bin/odyssey 12.19 123.4010.12 ------------------------------ --------------------------------------- Total 183.991592.75 8.66

  7. Sample Profile: Process Detail Elapsed Total Average Procedure Time (s)Energy (J)Power (W) ------------------------------ ---------------------------------------- _Dispatcher 0.25 2.53 10.11 _IOMGR_CheckDiscript 0.17 1.7410.23 _sftp_DataArrived 0.16 1.6810.48 _rpc2_RecvPacket 0.16 1.67 10.41 _ExaminePacket 0.16 1.66 10.35 . . . ------------------------------ --------------------------------------- Total 12.19 123.40 10.12

  8. Case Study: Adaptive Video Experimental Setup: Client: 75MHz 486 laptop Server: 200 MHz Pentium Pro Network: 900 MHz WaveLAN • Goals of case study: • Can degradation significantly reduce energy use? • Which degradations are most effective? • Are there techniques which reduce energy use across all video qualities?

  9. Result of Lossy Compression Benefit of lossy compression: 13% energy reduction

  10. Experiment: Display Size Reduction • Compression does not reduce X server energy usage. • Will reducing the display size help? • Repeated experiment with two additional tracks: • One each at highest and lowest compression. • Halved the height and width of the video display.

  11. Result of Reducing Display Size Cumulative benefit: 24% energy reduction

  12. Effect of Power Management • 94% of energy used to keep hardware in idle state. • Network power reduction: • Odyssey suspends network between RPCs. • Standby mode saves 1.4 W, overhead 0.81 ms. • Disk power management: • Odyssey puts disk to sleep at start of playback.

  13. Result of Power Management Overall result: 46% reduction in energy usage!

  14. Related Work • Low-Overhead CPU Profiling • Morph (Zhang, et al.) • DCPI (Anderson, et al.) • Network and Disk Power Measurement and Reduction • Network for Handheld Devices (Stemm and Katz) • Network Power Management (Kravets and Krishnan) • Disk (Douglis, Krishnan, and Marsh)

  15. Future Work • Only the beginning of the project: • Repeat case study for additional platforms. • Port to additional operating systems (Linux, …) . • Supplement analysis with histograms of energy usage over time. • Validate and calibrate with microbenchmarks. • Support for multiple concurrent applications.

  16. Summary PowerScope is a powerful tool for reducing energy usage. Adaptation reduces energy consumption. But, still a long way to go.

  17. System-Level Components • Viceroy: central point of resource control • monitors resources, notifies via upcall • also type-independent operations • Wardens: type-aware components • fidelity-changing operations • type-specific handling for efficiency

  18. Effect of Static Energy Usage

  19. Video Track Encodings Duration of each track is 184 seconds. Compression used: QuickTime / CinePak.

  20. IBM 701C Power Usage

More Related