Tweet
Tweet

Slide 1

Slide 1 text

GreenWeb: Language Extensions for Energy-Efficient Mobile Web Computing Yuhao Zhu The University of Texas at Austin with Vijay Janapa Reddi PLDI 2016 1

Slide 2

Slide 2 text

2 Web: Mobile Overtaking Desktop

Slide 3

Slide 3 text

0 30 60 90 120 2011 2012 2013 2014 2015 2016 2 Source: BIA/Kelsey Search Volume (B) Web: Mobile Overtaking Desktop

Slide 4

Slide 4 text

0 30 60 90 120 2011 2012 2013 2014 2015 2016 2 Source: BIA/Kelsey Search Volume (B) Mobile Desktop Web: Mobile Overtaking Desktop

Slide 5

Slide 5 text

0 30 60 90 120 2011 2012 2013 2014 2015 2016 2 Source: BIA/Kelsey Search Volume (B) Mobile Desktop Web: Mobile Overtaking Desktop

Slide 6

Slide 6 text

3 Web ≈ Mobile Web

Slide 7

Slide 7 text

Energy Concern Among Mobile Developers 4 [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 8

Slide 8 text

Energy Concern Among Mobile Developers 4 Percentage (%) 0 25 50 75 100 Mobile Desktop Data Center Never/Rarely Sometimes Often/Almost Always “My applications have requirements about energy usage.” [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 9

Slide 9 text

Energy Concern Among Mobile Developers 4 Percentage (%) 0 25 50 75 100 Mobile Desktop Data Center Never/Rarely Sometimes Often/Almost Always “My applications have requirements about energy usage.” [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 10

Slide 10 text

Energy Concern Among Mobile Developers 4 Percentage (%) 0 25 50 75 100 Mobile Desktop Data Center Never/Rarely Sometimes Often/Almost Always “My applications have requirements about energy usage.” [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 11

Slide 11 text

Developers are Willing to Make Trade-offs 5 [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 12

Slide 12 text

Developers are Willing to Make Trade-offs 5 Percentage (%) 0 25 50 75 100 Mobile Never/Rarely Sometimes Often/Almost Always “I'm willing to sacrifice performance, etc. for reduced energy usage.” [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 13

Slide 13 text

Developers are Willing to Make Trade-offs 5 Percentage (%) 0 25 50 75 100 Mobile Never/Rarely Sometimes Often/Almost Always “I'm willing to sacrifice performance, etc. for reduced energy usage.” [ICSE 2016] Manotas et al., “An Empirical Study of Practitioners’ Perspectives on Green Software Engineering”

Slide 14

Slide 14 text

Energy-efficiency 6

Slide 15

Slide 15 text

Quality-of-service Energy-efficiency 6

Slide 16

Slide 16 text

Quality-of-service Energy-efficiency Conflicting requirements 6

Slide 17

Slide 17 text

Quality-of-service Energy-efficiency Conflicting requirements 7 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing

Slide 18

Slide 18 text

8 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing

Slide 19

Slide 19 text

GreenWeb 8 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing

Slide 20

Slide 20 text

9 GreenWeb: Language for Energy-Efficiency ▸ Language abstractions for expressing QoS

Slide 21

Slide 21 text

9 ▸ Runtime that saves energy while meeting the QoS constraints GreenWeb: Language for Energy-Efficiency ▸ Language abstractions for expressing QoS

Slide 22

Slide 22 text

9 ▸ Runtime that saves energy while meeting the QoS constraints ▸ Result in 60% energy savings on real hardware/software implementations GreenWeb: Language for Energy-Efficiency ▸ Language abstractions for expressing QoS

Slide 23

Slide 23 text

9 ▸ Runtime the QoS constraints ▸ Result hardware/software implementations GreenWeb: Language for Energy-Efficiency ▸ Language abstractions for expressing QoS

Slide 24

Slide 24 text

10 What is QoS in mobile Web?

Slide 25

Slide 25 text

11 Understanding Mobile Web QoS

Slide 26

Slide 26 text

11 Performance QoS Experience Understanding Mobile Web QoS

Slide 27

Slide 27 text

11 Performance QoS Experience [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS

Slide 28

Slide 28 text

11 Performance QoS Experience [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS Too slow

Slide 29

Slide 29 text

11 Performance QoS Experience Unusable [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS Too slow

Slide 30

Slide 30 text

11 Performance QoS Experience Unusable Tolerable [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS Too slow

Slide 31

Slide 31 text

11 Performance QoS Experience Unusable Tolerable [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS Too slow Diminishing Returns

Slide 32

Slide 32 text

11 Performance QoS Experience Unusable Tolerable Imperceptible [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.” Understanding Mobile Web QoS Too slow Diminishing Returns

Slide 33

Slide 33 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 34

Slide 34 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 35

Slide 35 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 36

Slide 36 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 37

Slide 37 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS “Negative” Energy consumption Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 38

Slide 38 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 39

Slide 39 text

12 Performance QoS Experience Unusable Tolerable Imperceptible Understanding Mobile Web QoS Energy [OSDI 1996] Y. Endo et al., “Using Latency to Evaluate Interactive System Performance.”

Slide 40

Slide 40 text

13 Performance QoS Experience Unusable Tolerable Imperceptible Abstracting Mobile Web QoS

Slide 41

Slide 41 text

13 Performance QoS Experience Unusable Tolerable Imperceptible Abstracting Mobile Web QoS ▸ Performance metric ▹ Frame latency vs. Frame throughput

Slide 42

Slide 42 text

13 Performance QoS Experience Unusable Tolerable Imperceptible Abstracting Mobile Web QoS ▸ Performance metric ▹ Frame latency vs. Frame throughput QoS Type

Slide 43

Slide 43 text

13 Performance QoS Experience Unusable Tolerable Imperceptible Abstracting Mobile Web QoS ▸ Performance metric ▹ Frame latency vs. Frame throughput ▸ Threshold performance values ▹ Imperceptible target vs. Usable target QoS Type

Slide 44

Slide 44 text

13 Performance QoS Experience Unusable Tolerable Imperceptible Abstracting Mobile Web QoS ▸ Performance metric ▹ Frame latency vs. Frame throughput ▸ Threshold performance values ▹ Imperceptible target vs. Usable target QoS Type QoS Target

Slide 45

Slide 45 text

14 Expressing Mobile Web QoS

Slide 46

Slide 46 text

function animateMove() { /* Animation code omitted */ }
14 Expressing Mobile Web QoS

Slide 47

Slide 47 text

function animateMove() { /* Animation code omitted */ }
14 Expressing Mobile Web QoS element

Slide 48

Slide 48 text

function animateMove() { /* Animation code omitted */ }
14 Expressing Mobile Web QoS element event

Slide 49

Slide 49 text

function animateMove() { /* Animation code omitted */ }
14 Expressing Mobile Web QoS element event

Slide 50

Slide 50 text

function animateMove() { /* Animation code omitted */ }
14 Expressing Mobile Web QoS Expressing QoS at an event granularity element event

Slide 51

Slide 51 text

function animateMove() { /* Animation code omitted */ }

Slide 52

Slide 52 text

function animateMove() { /* Animation code omitted */ }

Slide 53

Slide 53 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 54

Slide 54 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 55

Slide 55 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 56

Slide 56 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 57

Slide 57 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 58

Slide 58 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 59

Slide 59 text

{ : Type, Target} function animateMove() { /* Animation code omitted */ }

Slide 60

Slide 60 text

16 Original application GreenWeb- annotated application GreenWeb Annotation Process Manual Annotation

Slide 61

Slide 61 text

16 Original application GreenWeb- annotated application GreenWeb Annotation Process Automatic Annotation?

Slide 62

Slide 62 text

16 Original application GreenWeb- annotated application GreenWeb Annotation Process Automatic Annotation? ▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations

Slide 63

Slide 63 text

16 GreenWeb- annotated application GreenWeb Annotation Process Automatic Annotation? ▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations DOM Tree

Slide 64

Slide 64 text

▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations 17 GreenWeb- annotated application GreenWeb Annotation Process

Slide 65

Slide 65 text

▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations 17 GreenWeb- annotated application GreenWeb Annotation Process Callback Instrumentation

Slide 66

Slide 66 text

▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations 17 GreenWeb- annotated application GreenWeb Annotation Process QoS Information Event Profiling Callback Instrumentation

Slide 67

Slide 67 text

▸ AutoGreen: automatically reasons about and inserts GreenWeb annotations 17 GreenWeb- annotated application GreenWeb Annotation Process QoS Information Event Profiling Annotation Generation Callback Instrumentation

Slide 68

Slide 68 text

▸ Language abstractions for expressing QoS 18 ▸ Runtime the QoS constraints ▸ Result hardware/software implementations GreenWeb: Language for Energy-Efficiency

Slide 69

Slide 69 text

▸ Language abstractions 18 ▸ Runtime that saves energy while meeting the QoS constraints ▸ Result hardware/software implementations GreenWeb: Language for Energy-Efficiency

Slide 70

Slide 70 text

19 GreenWeb Runtime Overview

Slide 71

Slide 71 text

19 GreenWeb Runtime Overview Frame Event

Slide 72

Slide 72 text

19 GreenWeb Runtime Overview Frame Event QoS Annotations

Slide 73

Slide 73 text

19 GreenWeb Runtime Overview Frame Event Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective QoS Annotations

Slide 74

Slide 74 text

QoS type: latency QoS target: 16 ms 19 GreenWeb Runtime Overview Frame Event Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 75

Slide 75 text

QoS type: latency QoS target: 16 ms 19 GreenWeb Runtime Overview Frame Event Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 76

Slide 76 text

QoS type: latency QoS target: 16 ms 19 GreenWeb Runtime Overview Frame Event Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 77

Slide 77 text

QoS type: latency QoS target: 16 ms 19 GreenWeb Runtime Overview Frame Event 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 78

Slide 78 text

QoS type: latency QoS target: 16 ms throughput 19 GreenWeb Runtime Overview Frame Event 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 79

Slide 79 text

QoS type: latency QoS target: 16 ms throughput 19 GreenWeb Runtime Overview Frame Event Frame Frame 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective

Slide 80

Slide 80 text

19 GreenWeb Runtime Overview Frame Event Frame Frame Time Event 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective QoS target: 2 s

Slide 81

Slide 81 text

19 GreenWeb Runtime Overview Frame Event Frame Frame Time Event Frame 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Runtime Objective 2 s QoS target: 2 s

Slide 82

Slide 82 text

19 GreenWeb Runtime Overview Frame Event Frame Frame Time Event Frame 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently { Runtime Objective 2 s QoS target: 2 s

Slide 83

Slide 83 text

19 GreenWeb Runtime Overview Frame Event Frame Frame Time Event Frame 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Frame Association { Runtime Objective 1 2 s

Slide 84

Slide 84 text

19 GreenWeb Runtime Overview Frame Event Frame Frame Time Event Frame 16 ms 16 ms 16 ms Enforcing event-level QoS at the frame-level energy-efficiently Frame Association Frame Scheduling { Runtime Objective 1 2 2 s

Slide 85

Slide 85 text

20 Frame Association Scripting Style Layout Paint Composite Frame Event

Slide 86

Slide 86 text

21 Frame Association S S L P C Frame Event

Slide 87

Slide 87 text

21 Frame Association S S L P C Frame Browser Process Renderer Process GPU Process Event

Slide 88

Slide 88 text

21 Frame Association S S L P C Frame Browser Process Renderer Process GPU Process Event Main Thread Compositor Thread

Slide 89

Slide 89 text

21 Frame Association S S L P C Frame Browser Process Renderer Process GPU Process Event Main Thread Compositor Thread IPC Inter-thread Message

Slide 90

Slide 90 text

22 Frame Association S S L P C Frame Browser Process Renderer Process GPU Process Event Frame S S L P C

Slide 91

Slide 91 text

22 Frame Association S S L P C Frame Browser Process Renderer Process GPU Process Event Frame S S L P C Distribute QoS information along with the communication messages

Slide 92

Slide 92 text

Choices of Energy-saving Techniques 23 GreenWeb can support a range of energy saving techniques

Slide 93

Slide 93 text

Choices of Energy-saving Techniques 23 GreenWeb can support a range of energy saving techniques ▹Dynamic resolution scaling [MobiCom 2015] ▹Power-saving display colors [MobiSys 2012] ▹Selective resource loading [NSDI 2015]

Slide 94

Slide 94 text

Choices of Energy-saving Techniques 23 GreenWeb can support a range of energy saving techniques ▹Dynamic resolution scaling [MobiCom 2015] ▹Power-saving display colors [MobiSys 2012] ▹Selective resource loading [NSDI 2015] ▹ACMP-based hardware mechanism

Slide 95

Slide 95 text

ACMP-based Hardware Substrate 24

Slide 96

Slide 96 text

ACMP-based Hardware Substrate 24 ▸ Asymmetric Chip-multiprocessor, a.k.a., Big/Little architecture

Slide 97

Slide 97 text

Energy Consumption Performance Big Core Small Core ACMP-based Hardware Substrate 24 ▸ Asymmetric Chip-multiprocessor, a.k.a., Big/Little architecture

Slide 98

Slide 98 text

Energy Consumption Performance Big Core Small Core ACMP-based Hardware Substrate 24 ▸ Asymmetric Chip-multiprocessor, a.k.a., Big/Little architecture Frequency Levels

Slide 99

Slide 99 text

Energy Consumption Performance Big Core Small Core ACMP-based Hardware Substrate 24 ▸ Asymmetric Chip-multiprocessor, a.k.a., Big/Little architecture ▸ Already used in commodity devices (e.g., Samsung Galaxy S6) Frequency Levels

Slide 100

Slide 100 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25

Slide 101

Slide 101 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25 ▸ Provide just enough energy to meet QoS constraints

Slide 102

Slide 102 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25 ▸ Provide just enough energy to meet QoS constraints div {ontouchend: latency, 16 ms}

Slide 103

Slide 103 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25 ▸ Provide just enough energy to meet QoS constraints 16 ms div {ontouchend: latency, 16 ms}

Slide 104

Slide 104 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25 ▸ Provide just enough energy to meet QoS constraints 16 ms div {ontouchend: latency, 16 ms}

Slide 105

Slide 105 text

Energy Consumption Performance Big Core Small Core ACMP-based GreenWeb Runtime 25 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015]

Slide 106

Slide 106 text

Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy =

Slide 107

Slide 107 text

Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = Tmemory + [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy =

Slide 108

Slide 108 text

Tcpu Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = Tmemory + [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy =

Slide 109

Slide 109 text

Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = Tmemory + Ncycles / f [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy =

Slide 110

Slide 110 text

Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = Tmemory + Ncycles / f [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy =

Slide 111

Slide 111 text

Energy Consumption Performance ACMP-based GreenWeb Runtime 26 ▸ Provide just enough energy to meet QoS constraints ▸ Event-based scheduling [HPCA 2015] Execution Time = Tmemory + Ncycles / f [PLDI 2003] Xie, et al., “Compile-Time Dynamic Voltage Scaling Settings: Opportunities and Limits” Energy = Execution Time x Power

Slide 112

Slide 112 text

▸ Language abstractions 27 ▸ Runtime that saves energy while meeting the QoS constraints ▸ Result hardware/software implementations GreenWeb: Language for Energy-Efficiency

Slide 113

Slide 113 text

▸ Language abstractions 27 ▸ Runtime the QoS constraints ▸ Result in 60% energy savings on real hardware/software implementations GreenWeb: Language for Energy-Efficiency

Slide 114

Slide 114 text

Real Hardware/Software Setup 28 ODroid XU+E development board, which contains an Exynos 5410 SoC used in Samsung Galaxy S4.

Slide 115

Slide 115 text

Real Hardware/Software Setup 28 ODroid XU+E development board, which contains an Exynos 5410 SoC used in Samsung Galaxy S4. Implementation incorporated into Chrome running on Android.

Slide 116

Slide 116 text

Real Hardware/Software Setup 28 ODroid XU+E development board, which contains an Exynos 5410 SoC used in Samsung Galaxy S4. Implementation incorporated into Chrome running on Android. UI-level record and replay for reproducibility. [ISPASS’15]

Slide 117

Slide 117 text

Evaluation ▸Baseline Mechanisms ▹Highest performance (Perf) — Standard to guarantee responsiveness ▹Interactive governor (Interactive) — Android default 29 29

Slide 118

Slide 118 text

Evaluation ▸Baseline Mechanisms ▹Highest performance (Perf) — Standard to guarantee responsiveness ▹Interactive governor (Interactive) — Android default 29 ▸Metrics ▹Energy Saving ▹QoS Violation 29

Slide 119

Slide 119 text

Evaluation ▸Baseline Mechanisms ▹Highest performance (Perf) — Standard to guarantee responsiveness ▹Interactive governor (Interactive) — Android default 29 ▸Metrics ▹Energy Saving ▹QoS Violation 29 ▸Applications ▹Top webpages (e.g., www.amazon.com) ▹Web Apps based on popular frameworks (e.g., Todo List)

Slide 120

Slide 120 text

30 Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf Evaluation Results

Slide 121

Slide 121 text

31 Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf Evaluation Results

Slide 122

Slide 122 text

32 Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf Evaluation Results

Slide 123

Slide 123 text

33 Evaluation Results QoS Violations (%) 0.0 0.8 1.5 2.3 3.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf

Slide 124

Slide 124 text

Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf 34 Evaluation Results QoS Violations (%) 0.0 0.8 1.5 2.3 3.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN No QoS Violations

Slide 125

Slide 125 text

Norm. Energy 0.0 0.3 0.5 0.8 1.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN GreenWeb Interactive Perf 34 Evaluation Results QoS Violations (%) 0.0 0.8 1.5 2.3 3.0 CamanJS Craigslist Paperjs Goo Google Todo CNet BBC LZMA-JS Amazon W3School MSN 29.2% - 66.0% energy savings, 0.8% more QoS violations No QoS Violations

Slide 126

Slide 126 text

35 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing

Slide 127

Slide 127 text

35 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing Abstraction Express QoS constraints

Slide 128

Slide 128 text

35 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing Abstraction Express QoS constraints Runtime Satisfy QoS specifications using energy saving techniques

Slide 129

Slide 129 text

35 GreenWeb Programming language support for balancing energy-efficiency and QoS in mobile Web computing Abstraction Express QoS constraints Runtime Satisfy QoS specifications using energy saving techniques Effect Significant energy savings

Slide 130

Slide 130 text

36 wattwiseweb.org

Slide 131

Slide 131 text

37 GreenWeb: Language Extensions for Energy-Efficient Mobile Web Computing Yuhao Zhu The University of Texas at Austin with Vijay Janapa Reddi PLDI 2016