$30 off During Our Annual Pro Sale. View Details »

A Microservice Architecture for Online Mobile App Optimization

Yixue Zhao
May 25, 2019
Research
1
370

A Microservice Architecture for Online Mobile App Optimization

Presentation slides of the paper "A Microservice Architecture for Online Mobile App Optimization" at the International Conference on Mobile Software Engineering and Systems (MOBILESoft) 2019.

The presentation can be found at
Pre-print can be found at https://arxiv.org/abs/1902.08879

Yixue Zhao

May 25, 2019
Tweet

More Decks by Yixue Zhao

See All by Yixue Zhao
Avgust: Automating Usage-Based Test Generation from Videos of App Executions
yixue_zhao
0
46
Best Dissertation Symposium: Reducing user-perceived latency in mobile applications via prefetching and caching
yixue_zhao
0
210
Assessing the Feasibility of Web-Request Prediction Models on Mobile Platforms
yixue_zhao
0
290
FrUITeR: A Framework for Evaluating UI Test Reuse
yixue_zhao
0
390
FrUITeR: A Framework for Evaluating UI Test Reuse (Trailer Video Slides)
yixue_zhao
0
320
Empirically Assessing Opportunities for Prefetching and Caching in Mobile Apps
yixue_zhao
0
280
Leveraging Program Analysis to Reduce User-Perceived Latency in Mobile Applications
yixue_zhao
0
770

Other Decks in Research

See All in Research
強い生成 AI 検知システムを実現する Data-Centric なデータセット管理
johannyjm
1
680
阪神タイガース優勝のひみつ - Pythonでシュッと調べた件 / SABRmetrics for Python
shinyorke
1
280
20230923_Kaggler_meetup_LT_ya9do.pdf
ya9do
2
1.3k
近似最近傍探索とVector DBの理論的背景
matsui_528
2
310
Active Retrieval Augmented Generation
kiyohiro8
3
270
マルチリンガルな言語モデル入門：これまでとこれから
ryou0634
2
2.1k
論文紹介：What In-Context Learning “Learns” In-Context: Disentangling Task Recognition and Task Learning
yuri00
0
270
訪問介護におけるシフトスケジューリングモデルと自社データによる検証 / orsj 2023 fall
sms_tech
0
290
遺伝的アルゴリズムの簡単な紹介
kentaitakura
0
320
Network Pruning：枝刈り手法のサーベイ
kohamax0724
2
1.1k
Coco-Nut: 自由記述文による声質制御に向けた多話者音声・声質自由記述ペアデータセット
aya_watanabe
0
210
SNLP2023: From Characters to Words: Hierarchical Pre-trained Language Model for Open-vocabulary Language Understanding
tathi
0
170

Featured

See All Featured
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
10
1.3k
In The Pink: A Labor of Love
frogandcode
135
21k
Happy Clients
brianwarren
91
6.1k
"I'm Feeling Lucky" - Building Great Search Experiences for Today's Users (#IAC19)
danielanewman
219
21k
A designer walks into a library…
pauljervisheath
199
22k
The Cost Of JavaScript in 2023
addyosmani
10
3.1k
The World Runs on Bad Software
bkeepers
PRO
59
6.2k
Scaling GitHub
holman
455
140k
How GitHub (no longer) Works
holman
299
140k
Side Projects
sachag
450
39k
Art Directing for the Web. Five minutes with CSS Template Areas
malarkey
197
11k
Faster Mobile Websites
deanohume
296
29k

Transcript

  1. 1
    A Microservice Architecture for Online
    Mobile App Optimization
    Yixue Zhao, Nenad Medvidović
    University of Southern California
    May 25, 2019

    View Slide

  2. 2
    Background

    View Slide

  3. 3
    Background
    Auxiliary
    e.g., empirical
    study
    Detection
    e.g., security
    vulnerabilities
    Optimization
    e.g., bug
    repair

    View Slide

  4. 4
    Background
    Limited Optimization Techniques
    • 48 papers (mobile domain) accepted at
    ICSE (2014 – 2018)
    • only 7 optimization techniques
    • Limited evaluation
    e.g., 5 apps

    View Slide

  5. 5
    Motivation
    • Angry users want better apps
    • Developers want better apps
    • Researchers need to propose more
    optimization techniques to address real needs

    View Slide

  6. 6
    Common Challenges
    • “It’s hard to use other research tools”
    • “They don’t have documentation”
    • “It’s so time-consuming to find subject apps”
    • “It’s hard to debug a research tool on real
    apps without domain-specific knowledge”
    • …

    View Slide

  7. 7
    Problem Summary
    • Reusability issue of existing tools
    • Gap between researchers and developers

    View Slide

  8. 8
    Our Solution
    • Reusability issue of existing tools
    MAOMAO (reference architecture)
    • Gap between researchers and developers
    Online Repository

    View Slide

  9. 9
    • Case Study
    • MAOMAO
    • MAOMAO Repository
    Outline

    View Slide

  10. 10
    Case Study: PALOMA
    • Leveraging Program Analysis to Reduce User-
    Perceived Latency in Mobile Applications
    (ICSE 2018)
    Static
    Analysis
    App
    Instrumentation
    Optimized
    App

    View Slide

  11. 11
    PALOMA Workflow
    Static Analysis App
    Instrumenter Optimized App
    String
    Analyzer
    Callback
    Analyzer
    Original
    App
    Proxy
    Runtime
    Prefetching
    Origin
    Server
    offline online

    View Slide

  12. 12
    PALOMA Reuse Opportunities
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    Violist (FSE 2015) Gator (ICSE 2015)
    PALOMA
    Proxy
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  13. 13
    PALOMA Reuse Opportunities
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    Violist (FSE 2015) Gator (ICSE 2015)
    PALOMA
    Proxy
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  14. 14
    PALOMA Reuse Opportunities
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    Violist (FSE 2015) Gator (ICSE 2015)
    PALOMA
    Proxy
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  15. 15
    PALOMA Reuse Opportunities
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  16. 16
    PALOMA Reuse Opportunities
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  17. 17
    Can we do it better?
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    Reflection
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  18. 18
    Can we do it better?
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  19. 19
    Let’s do it better!
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    Static
    Analyzer
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()

    View Slide

  20. 20
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    Static
    Analyzer
    Static
    Analyzer
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()
    Let’s do it better!

    View Slide

  21. 21
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    Static
    Analyzer
    Static
    Analyzer
    App
    Instrumenter
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()
    Let’s do it better!

    View Slide

  22. 22
    String
    Analyzer
    Callback
    Analyzer
    App
    Instrumenter
    PALOMA
    Proxy
    Violist (FSE 2015) Gator (ICSE 2015)
    e.g., Insert_TriggerPrefetch() e.g., TriggerPrefetch()
    Static
    Analyzer
    Static
    Analyzer
    App
    Instrumenter
    Backend
    Service
    Let’s do it better!

    View Slide

  23. 23
    • A Microservice Architecture for Online
    Mobile App Optimization
    • Light-weight, black-box microservices
    • Easy to reuse and adopt
    • Suitable for heterogeneous mobile domain
    e.g., different languages, technologies
    Let’s do it better with MAOMAO

    View Slide

  24. 24
    MAOMAO Reference Architecture
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter

    View Slide

  25. 25
    MAOMAO Reference Architecture
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter

    View Slide

  26. 26
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter
    MAOMAO Reference Architecture

    View Slide

  27. 27
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter
    MAOMAO Reference Architecture

    View Slide

  28. 28
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter
    MAOMAO Reference Architecture

    View Slide

  29. 29
    MAOMAO Reference Architecture
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter

    View Slide

  30. 30
    Other Examples
    New
    Techniques?

    View Slide

  31. 31
    MAOMAO Benefits
    • Loosely-coupled modular design
    • Easy to update individual components
    • Easy to reuse for researchers
    • Easy to adopt by developers
    • …
    Better Apps!

    View Slide

  32. 32
    MAOMAO Repository
    Researchers
    App Developers
    Open Repository
    Microservice
    Pool
    Service Request
    Pool
    Benchmark
    Pool
    Test Script
    Pool
    Testing Environment
    Testing
    Engine
    Portal
    test results

    View Slide

  33. 33
    MAOMAO Repository
    Open Science
    Community

    View Slide

  34. 34
    Our Vision
    We need you!

    View Slide

  35. 35
    Want to know more? J
    Contact: Yixue Zhao ([email protected])
    Follow-up Paper: Mobile-App Analysis and Instrumentation
    Techniques Reimagined with DECREE (ICSE 2019, Doctoral Symposium)
    Follow-up Talk: May 30 Thursday, 12:57pm, Agora lobby (ROSE Festival)

    View Slide

  36. 36

    View Slide

  37. 37

    View Slide

  38. 38
    Legend
    Intermediate Representation
    MAOMAO’s Components
    Program Point
    Configuration
    REST API
    API Documentation
    General Data
    Original App
    Instrumented App
    Offline App Development
    Static
    Analyzer
    Backend
    Service
    Instrumented OS
    Online App Runtime
    Original OS
    Intermediate
    Representer
    Device
    Monitor
    OS
    Instrumenter
    App
    Instrumenter

    View Slide