Event-Based Scheduling for Energy-Efficient Quality of Service (eQoS) in Mobile Web Applications

HPCA 2015

Transcript

1. Event-Based Scheduling for Energy-Efficient Quality of Service (eQoS) in Mobile Web Applications Yuhao Zhu, Matthew Halpern, Vijay Janapa Reddi Department of Electrical and Computer Engineering The University of Texas at Austin HPCA — Feb. 9th, 2015
2. Responsiveness
3. Responsiveness ▸100 ms latency is the limit for having users feel the system is reacting responsively [Miller 1968; Card et al. 1991]
4. Responsiveness ▸100 ms latency is the limit for having users feel the system is reacting responsively [Miller 1968; Card et al. 1991] ▸64% of mobile users will not revisit a slow Web app [Source: Akamai]
5. Responsiveness ▸100 ms latency is the limit for having users feel the system is reacting responsively [Miller 1968; Card et al. 1991] ▸64% of mobile users will not revisit a slow Web app [Source: Akamai] ▸Amazon found every 100 ms of latency costs them 1% in sales per year [Source: Amazon]
6. Responsiveness Energy-Efficiency
7. Responsiveness Energy-Efficiency Conflicting requirements
8. eQoS: Making a Calculated Trade-off 3 Performance degradation QoS Experience
9. eQoS: Making a Calculated Trade-off 3 Performance degradation QoS Experience Imperceptible
10. eQoS: Making a Calculated Trade-off 3 Performance degradation QoS Experience Imperceptible Tolerable
11. eQoS: Making a Calculated Trade-off 3 Performance degradation QoS Experience Imperceptible Tolerable Unusable
12. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
13. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
14. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
15. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
16. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
17. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
18. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
19. eQoS: Making a Calculated Trade-off 3 Energy Savings Performance degradation QoS Experience Imperceptible Tolerable Unusable
20. Interacting with an App 4
21. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4
22. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4
23. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area
24. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing Interactions
25. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing Interactions
26. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing pressing Interactions
27. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing pressing Interactions
28. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing pressing moving Interactions
29. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing pressing moving Interactions Events
30. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area typing pressing moving Interactions Events keypress click touchmove
31. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area Events keypress click touchmove
32. Interacting with an App 4 http://yuhaozhu.com/googlemaps.mp4 bay area Events keypress click touchmove 1487 Events
33. Mobile Web Applications are Event-Driven 5
34. Mobile Web Applications are Event-Driven 5 Event FIFO Queue
35. Mobile Web Applications are Event-Driven 5 Event FIFO Queue Head Tail
36. Mobile Web Applications are Event-Driven 5 Touch Screen Event Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue
37. Mobile Web Applications are Event-Driven 5 Touch Screen Event Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue Event Loop
38. Mobile Web Applications are Event-Driven 5 Touch Screen Event Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue Fetch for Execution Event Loop
39. Mobile Web Applications are Event-Driven 5 Touch Screen Event Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue Fetch for Execution Event Loop
40. Mobile Web Applications are Event-Driven 5 Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue Fetch for Execution Event Loop
41. Mobile Web Applications are Event-Driven 5 Event FIFO Queue Key Press Event … … Head Tail Timer Event Insert to Event Queue Fetch for Execution Event Loop Optimize for eQoS at an event-granularity
42. ▸ Key idea: Execute events with just-enough energy to meet user QoS expectation ▹ Develop a runtime system that leverages the large scheduling space of heterogeneous CPUs 6 Executive Summary
43. ▸ Key idea: Execute events with just-enough energy to meet user QoS expectation ▹ Develop a runtime system that leverages the large scheduling space of heterogeneous CPUs 6 Executive Summary Event-level Characterization
44. ▸ Key idea: Execute events with just-enough energy to meet user QoS expectation ▹ Develop a runtime system that leverages the large scheduling space of heterogeneous CPUs Runtime Mechanics to Exploit Latency Slack 6 Executive Summary Event-level Characterization
45. ▸ Key idea: Execute events with just-enough energy to meet user QoS expectation ▹ Develop a runtime system that leverages the large scheduling space of heterogeneous CPUs Runtime Mechanics to Exploit Latency Slack Heterogeneous Resource Utilization 6 Executive Summary Event-level Characterization
46. ▸ Key idea: Execute events with just-enough energy to meet user QoS expectation ▹ Develop a runtime system that leverages the large scheduling space of heterogeneous CPUs Runtime Mechanics to Exploit Latency Slack Heterogeneous Resource Utilization 6 Executive Summary Event-level Characterization
47. Exploit the Slack in Events 7 150 100 50 0 Event Latency (ms) Events
48. Exploit the Slack in Events 8 150 100 50 0 Event Latency (ms) Events
49. Exploit the Slack in Events 8 Unusable QoS 150 100 50 0 Event Latency (ms) Events
50. Exploit the Slack in Events 8 keyup Unusable QoS 150 100 50 0 Event Latency (ms) Events
51. Exploit the Slack in Events 8 keyup Unusable QoS 150 100 50 0 Event Latency (ms) Events
52. Exploit the Slack in Events 8 keyup Large Slack Unusable QoS 150 100 50 0 Event Latency (ms) Events
53. Exploit the Slack in Events 8 keyup Large Slack change Unusable QoS 150 100 50 0 Event Latency (ms) Events
54. Exploit the Slack in Events 8 keyup Large Slack change Small Slack Unusable QoS 150 100 50 0 Event Latency (ms) Events
55. Exploit the Slack in Events 8 keyup Large Slack change Small Slack click Unusable QoS 150 100 50 0 Event Latency (ms) Events
56. Exploit the Slack in Events 8 keyup Large Slack change Small Slack click Unusable QoS ▸ Wide distribution of event latencies. Events exhibit different slacks. ▹ How to exploit different slacks for different events? 150 100 50 0 Event Latency (ms) Events
57. Event-Based Scheduler (EBS) Overview 9 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event
58. Event-Based Scheduler (EBS) Overview 9 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Thread Scheduling
59. Event-Based Scheduler (EBS) Overview 9 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Thread Scheduling
60. Event-Based Scheduler (EBS) Overview 9 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Thread-based Scheduler Thread Scheduling
61. Event-Based Scheduler (EBS) Overview 9 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Thread-based Scheduler Thread Scheduling Throughput Fairness
62. Event-Based Scheduler (EBS) Overview 10 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event
63. Event-Based Scheduler (EBS) Overview 10 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Mobile applications
64. Event-Based Scheduler (EBS) Overview 10 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Mobile applications Event Queue
65. Event-Based Scheduler (EBS) Overview 10 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Event-based Scheduler Mobile applications Event Queue
66. Event-Based Scheduler (EBS) Overview 10 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event Event-based Scheduler Mobile applications eQoS Event Queue
67. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Event Queue
68. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Event Queue Big/Little Hardware
69. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Event Queue Big/Little Hardware Event-Based Scheduler
70. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Event Queue Big/Little Hardware Event-Based Scheduler Event
71. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Event Queue Big/Little Hardware Event-Based Scheduler <core, freq> Event
72. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup QoS Monitor Event Queue Big/Little Hardware Event-Based Scheduler <core, freq> Event
73. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup QoS Monitor Event Queue Big/Little Hardware Event-Based Scheduler Model <core, freq> Event
74. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup QoS Monitor Model Constructor Event Queue Big/Little Hardware Event-Based Scheduler Model <core, freq> Event
75. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup QoS Monitor Model Constructor Event Queue Big/Little Hardware Event-Based Scheduler Model <core, freq> Event
76. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Detector QoS Monitor Model Constructor Event Queue Big/Little Hardware Event-Based Scheduler Model <core, freq> Event
77. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Detector QoS Monitor Model Constructor Event Queue Big/Little Hardware Event-Based Scheduler Model <core, freq> Event
78. Event-Based Scheduler (EBS) Overview 11 ▸ Goal: Find the most energy-efficient architectural configuration that meets the QoS target for each event H3 … H2 … H1 change click keyup Detector QoS Monitor Model Constructor Event Queue Big/Little Hardware Event-Based Scheduler Model Recalibrate <core, freq> Event
79. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config
80. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Memory Operation CPU Operation Tmemory Ndependent f Event Latency Xie, et al., Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits, PLDI’03
81. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Memory Operation CPU Operation Tmemory Ndependent f Event Latency Xie, et al., Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits, PLDI’03 Event Latency =
82. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Memory Operation CPU Operation Tmemory Ndependent f Event Latency Xie, et al., Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits, PLDI’03 Event Latency = Tmemory +
83. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Memory Operation CPU Operation Tmemory Ndependent f Event Latency Xie, et al., Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits, PLDI’03 Event Latency = Tmemory + Ndependent / f
84. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Memory Operation CPU Operation Tmemory Ndependent f Event Latency Xie, et al., Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits, PLDI’03 Event Latency = Tmemory + Ndependent / f
85. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Event Latency = Tmemory + Ndependent / f Event Latency Frequency
86. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the Model Constructor Works 12 ▸ Goal: Estimate the event latency under each <core, freq> config Event Latency = Tmemory + Ndependent / f Event Latency Frequency
87. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Feedback
88. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Event Latency (s) Event Intensity (evt/s) Feedback
89. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Event Latency (s) Event Intensity (evt/s) Feedback
90. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Event Latency (s) Event Intensity (evt/s) Feedback
91. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Event Latency (s) Event Intensity (evt/s) NA Feedback
92. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> How the Detector Works 13 ▸ Goal: Find the QoS target for each event CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 CamanJS Pdf.js Crypto Zlib Paper.js Ember.js GWT Backbone jQuery sina google ebay Doom Rain 10-4 104 106 100 Event Latency (s) Event Intensity (evt/s) NA Feedback [1, 10] s [50, 100] ms [60, 30] fps
93. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 14 ▸ Goal: Predict the <core, frequency> for the next event
94. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 14 ▸ Goal: Predict the <core, frequency> for the next event Predicted Latency Frequency Big Core Little Core
95. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 14 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core
96. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 14 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core
97. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 14 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core ▸ Scheduling overhead (~120 us): ▹ Core migration ▹ Frequency switching
98. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 15 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core
99. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 15 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core
100. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 15 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core ▸ Fine-tune the execution upon over- prediction or under-prediction
101. Feedback Detector QoS Monitor Model Constructor Event Info Model Recalibrate <core, freq> Feedback How the QoS Monitor Works 15 ▸ Goal: Predict the <core, frequency> for the next event QoS Target Predicted Latency Frequency Big Core Little Core ▸ Fine-tune the execution upon over- prediction or under-prediction ▸ Recalibrate if it mispredicts too often
102. Evaluation Methodology ▸ Implemented inside Google Chromium Web browser ▸ Representative hardware platform ▹ Exynos 5410 SoC (A15 + A7) 16
103. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17
104. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17
105. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Scheduling Scenarios
106. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Scheduling Scenarios ▹ Scheduling for imperceptibility
107. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Scheduling Scenarios ▹ Scheduling for imperceptibility ▹ Scheduling for tolerability
108. Evaluation Methodology ▸ Baseline Mechanisms ▹ Highest performance (Perf) — Standard to guarantee responsiveness ▹ Minimal energy (Energy) — Minimize energy consumption ▹ Interactive governor (Interactive) — Android default ▹ On-demand governor (Ondemand) 17 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Scheduling Scenarios ▹ Scheduling for imperceptibility ▹ Scheduling for tolerability
109. Evaluation Results 18 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Ondemand Energy
110. 19 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy Evaluation Results No QoS Violations
111. 20 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy Evaluation Results No QoS Violations
112. 21 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9 Evaluation Results
113. 22 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9 Evaluation Results
114. 23 QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9 Evaluation Results Energy (J) 0.0 1.0 2.0 3.0 4.0 emberjs gwt jquery backbone paperjs sina google ebay
115. 24 Energy (J) 0.0 1.0 2.0 3.0 4.0 emberjs gwt jquery backbone paperjs sina google ebay 8.2 7.7 Evaluation Results QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9
116. 25 Energy (J) 0.0 1.0 2.0 3.0 4.0 emberjs gwt jquery backbone paperjs sina google ebay 8.2 7.7 Evaluation Results QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9
117. 26 Energy (J) 0.0 1.0 2.0 3.0 4.0 emberjs gwt jquery backbone paperjs sina google ebay 8.2 7.7 Evaluation Results QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9
118. 26 Energy (J) 0.0 1.0 2.0 3.0 4.0 emberjs gwt jquery backbone paperjs sina google ebay 8.2 7.7 Evaluation Results QoS Violations (%) 0.0 1.5 3.0 4.5 6.0 emberjs gwt jquery backbone paperjs sina google ebay EBS Perf Interactive Energy 9.4 17.8 58.1 6.9 37.9% - 41.2% energy savings, 0.1% more QoS violations
119. What Have We Learnt So Far 27 ▸ User interactivity and human perceptibility are important because it has economic ramifications and impacts user device choice, as well as end-user satisfaction
120. What Have We Learnt So Far 27 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Need a systematic way of understanding and QoS-energy trade-offs. To that end we advocate eQoS to strike a balance between user satisfaction and energy reduction ▸ User interactivity and human perceptibility are important because it has economic ramifications and impacts user device choice, as well as end-user satisfaction
121. What Have We Learnt So Far 27 PH Imperceptible 0 PI PU PL QoSU QoSI Tolerable Unusable -ESU ESI EST Quality-of-Service (QoS) Energy Savings (ES) Performance Degradation ▸ Need a systematic way of understanding and QoS-energy trade-offs. To that end we advocate eQoS to strike a balance between user satisfaction and energy reduction ▸ Develop the event-based scheduling (EBS) mechanism that exploits the fundamental event-driven execution pattern of mobile applications to achieve better eQoS ▸ User interactivity and human perceptibility are important because it has economic ramifications and impacts user device choice, as well as end-user satisfaction Heterogeneous Hardware Event Queue H3 … H2 … H1 Dispatch <Core, Freq> PI, PU Detector QoS Monitor Model Constructor Recalibrate Event-Based Scheduler Event Info Models onkeyup=“H1 () {…}” keyup onchange=“H2 () {…}” change Event execution feedback onclick=“H3 () {…}” click
122. Mobile is One Example 28 Web Mobile
123. Mobile is One Example 28 Web Mobile Internet-of-Things Sensor networks Cloud
124. Mobile is One Example 28 Web Mobile Internet-of-Things Sensor networks Cloud Event-based processing is a fundamental computation pattern.
125. Mobile is One Example 28 Web Mobile Internet-of-Things Sensor networks Cloud Event-based processing is a fundamental computation pattern.
126. ▸ Another talk @ Best of CAL session ▸ Wednesday after the keynote ▸ How computer architects can improve Web technologies
127. Thank you