1 / 31

PICSEL: Measuring User-Perceived Performance to Control Dynamic Frequency Scaling

Arindam Mallik Jack Cosgrove Robert P. Dick Gokhan Memik Peter Dinda Northwestern University Department of Electrical Engineering and Computer Science Evanston, Illinois, USA. PICSEL: Measuring User-Perceived Performance to Control Dynamic Frequency Scaling. Claims.

leif
Download Presentation

PICSEL: Measuring User-Perceived Performance to Control Dynamic Frequency Scaling

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. Arindam Mallik Jack Cosgrove Robert P. Dick Gokhan Memik Peter Dinda Northwestern University Department of Electrical Engineering and Computer Science Evanston, Illinois, USA PICSEL: Measuring User-Perceived Performance to Control Dynamic Frequency Scaling ASPLOS • March 3, 2008 • Seattle, Washington, USA 1

  2. Claims • Traditional performance metrics do not measure user-perceived performance well • Our performance metrics measure user-perceived performance better • PICSEL is a power management policy that uses our metrics to achieve system power improvements of up to 12.1% compared to existing policies 2

  3. PICSEL Overview Main Memory Display Screenshot Compare consecutive screenshots Redraw screen Change frequency CPU 3

  4. PICSEL Overview Main Memory Display Screenshot Compare consecutive screenshots “The ultimate goal of a computer system is to satisfy the user” Redraw screen Change frequency CPU 4

  5. Summary • Power problem • DVFS • System performance • Traditional vs. user-perceived • PICSEL • How it works • Results • Conclusions 5

  6. Need for Power Management • Energy-hungry processors present three major problems: • Higher energy consumption • Shorter battery life • Higher temperatures 6

  7. DVFS Is a Potential Solution • Dynamic voltage and frequency scaling (DVFS) addresses all three problems • Trades off processor frequency for energy savings • Commonly used • Ideal DVFS policy: Find the lowest level of performance acceptable to the user to maximize power savings 7

  8. How Fast Is Fast Enough? • Human in loop is often rate-limiter Output Devices (kHz) User (Hz) Processor (GHz) Input Devices (kHz) 8

  9. Traditional Performance • Traditional performance metrics focus on processor performance • “Close to metal” Output Devices Processor (IPS) User Input Devices 9

  10. User-Perceived Performance • User-perceived performance metrics focus on interface device performance • “Close to flesh” Output Devices (Display, Speakers) User (N/A) Processor Input Devices (Mouse, Keyboard) 10

  11. Display Is Rich in Data • Use change in pixel intensities as metric for user-perceived performance 11

  12. PICSEL Perception Informed CPU performance Scaling to Extend battery Life 12

  13. Measure Pixel Intensities • Windows GDI Screenshot • Capture contiguous area of screen • Repeat periodically • Compare RGB intensities across samples Cached Ri Ri-1 RΔ Gi Gi-1 G Δ - = Bi Bi-1 B Δ 13

  14. Performance Measurements • Average Pixel Change (APC) • APC = (RΔ + GΔ + BΔ) / 3 • Averaged across all pixels • Measures “slowness” of display • Rate of Average Pixel Change (APR) • APR = (APCi – APCi-1)/(Ti – Ti-1) • Measures “jitter” of display 14

  15. Cost of PICSEL • PICSEL uses <2% CPU utilization • Cost of target applications is 50-100% CPU utilization 15

  16. APC and IPS Perform Differently 16

  17. PICSEL Keeps Performance Within Bounds “No change” band APC Increase frequency “No change” band APR Make a decision on these marks Time 17

  18. PICSEL Algorithm IF (APCinit - μAPC) < ρ ×(1-α) × APCinit OR |APRinit- μAPR| < γ ×(1-α) × APRinit Reduce f by one level Reset α of the last level to 0.0 ELSE Increase f by one level Increment α by 0.1 18

  19. Two Versions of PICSEL • All values chosen by authors after testing using target applications • Too long (243 days) to construct ideal values • User evaluation “closed the loop” 19

  20. PICSEL Evaluation • 20 users • Shockwave animation and DVD movie play for 2 minutes • FIFA game plays for 3.5 minutes • Three randomly selected trials per application • One double-blind DVFS policy for each trial • User rates satisfaction from one (lowest) to five (highest) after each trial 20

  21. Shockwave Frequency Trace 21

  22. DVD Frequency Trace 22

  23. Game Frequency Trace 23

  24. Results * Not Different with 95% confidence ** Different with 90% confidence 24

  25. Results * Not Different with 95% confidence ** Different with 90% confidence 25

  26. Thermal Emergencies at Maximum Frequency under Windows DVFS 26

  27. Thermal Emergencies Hurt User-Perceived Performance Perceived slowdown 27

  28. PICSEL Reduces Thermal Emergencies • User satisfaction is maximized by cPICSEL • Frequency is high enough to deliver good performance but not high enough to trigger thermal emergencies 28

  29. Conclusions • Display performance is a better metric for controlling DVFS than processor performance • Existing processor performance-based DVFS policies have slack that can be exploited • Cost of monitoring the display output is low • User satisfaction is the same or better 29

  30. Related DVFS Work • Based on GUI events • Gurun, S. and Krintz, C. 2005. AutoDVS: an Automatic, General-purpose, Dynamic Clock Scheduling System for Hand-held Devices. In Proc. of the 5th ACM Int. Conf. on Embedded Software (EMSOFT’05), 218-226. • Based on application messages • Flautner, K. and Mudge, T. 2002. Vertigo: Automatic Performance-Setting for Linux. ACM SIGOPS Operating Systems Review 36, SI (Winter 2002), 105-116. 30

  31. Thank You! Check out “Empathic Computer Architectures and Systems” at Wild and Crazy Ideas and visit empathicsystems.org for more user-centered systems research 31

More Related