Collaborative Detection of Energy Bugs

Slides:



Advertisements
Similar presentations
Android Power Calculations Approaches and Best Practice Hafed Alghamdi.
Advertisements

Fine-Grained Power Modeling for Smartphones Using System Call Tracing Abhinav Pathak, Y. Charlie Hu Purdue University Ming Zhang, Paramvir Bahl, Yi-Min.
Sophos Mobile Control. Tablets on the rise 2 Trends 3 75% of 157 polled companies encourage employee owned smart phones and tablets to access corporate.
Understanding Human-Smartphone Concerns: A Study of Battery Life Denzil Ferreira, Anind K. Dey, Vassilis Kostakos Pervasive 2011.
Multi-criteria infrastructure for location-based applications Shortly known as: Localization Platform Ronen Abraham Ido Cohen Yuval Efrati Tomer Sole'
Statistical Treatment of Data Significant Figures : number of digits know with certainty + the first in doubt. Rounding off: use the same number of significant.
Accelerating innovation in home technology Ratul Mahajan.
ThinkAir: Dynamic Resource Allocation and Parallel Execution in Cloud for Mobile Code Offloading Sokol Kosta, Pan Hui Deutsche Telekom Labs, Berlin, Germany.
Case study 2 Android – Mobile OS.
WuKong: Automatically Detecting and Localizing Bugs that Manifest at Large System Scales Bowen ZhouJonathan Too Milind KulkarniSaurabh Bagchi Purdue University.
 Energy Results: Memory Assistant Arcade Game  Performance Results:  Response Time ▪ Memory assistant: 17.3 sec -> 1.5 sec ▪ Arcade game: 6 FPS -> 13.
Characterizing and Modeling the Impact of Wireless Signal Strength on Smartphone Battery Drain Ning Ding Xiaomeng Chen Abhinav Pathak Y. Charlie Hu 1 Daniel.
Smart Phone Laboratory ECEN 489 Srinivas Shakkottai.
ReCapture A Pattern-aware Benchmark Tool for Smartphones.
Leave Me Alone: App-level Protection Against
AppInsight: Mobile App Performance Monitoring In The Wild Lenin Ravindranath, Jitu Padhye, Sharad Agarwal, Ratul Mahajan, Ian Obermiller, Shahin Shayandeh.
Embedded System Lab. 정범종 A_DRM: Architecture-aware Distributed Resource Management of Virtualized Clusters H. Wang et al. VEE, 2015.
© 2015 AT&T Intellectual Property. All rights reserved. AT&T, the AT&T logo and all other marks contained herein are trademarks of AT&T Intellectual Property.
Ivan Vitev & The First Precise Determination of Quark Energy Loss in Nuclei Ivan Vitev (PI), Ming Liu (Co-PI), Patrick McGaughey, Benwei Zhang T-16 and.
By: Chisha Malama Supervisor: Dr WD Tucker Co Supervisor: Mr M. Norman.
Mobile Testing Overview. Agenda Mobile application quality poses a unique challenge Mobile changes the ALM cycle – Interoperability is unique to mobile.
Power Guru: Implementing Smart Power Management on the Android Platform Written by Raef Mchaymech.
Best detection scheme achieves 100% hit detection with
Efficient Opportunistic Sensing using Mobile Collaborative Platform MOSDEN.
A Software Energy Analysis Method using Executable UML for Smartphones Kenji Hisazumi System LSI Research Center Kyushu University.
Mobile App Upgrade and Migration Process mobile.semaphore-software.com.
Announce-1 CSE 5810Announcements  Informatics is:  Management and Processing of Data  From Multiple Sources/Contexts  Involves Classification (Ontologies),
The Fallacy Behind “There’s Nothing to Hide” Why End-to-End Encryption Is a Must in Today’s World.
Mary Ganesan and Lora Strother Campus Tours Using a Mobile Device.
Smartphone energy considerations (for browser design) Ratul Mahajan Microsoft Research.
Smartphone Energy Drain in the Wild: Analysis and Implications Authors: Xiaomeng Chen, Ning Ding, Abhilash Jindal†, Y. Charlie Hu†, Maruti Gupta, Rath.
By: Amol Kokje Tosha Shah Raymond Tyler. Outline of Presentation Motivation Goals Methodology Application Flow What we have done To do Possible extensions.
WELCOME Mobile Applications Testing
Software Engineering in Mobile Application Development
Mobile Testing - Bug Report
Chapter Objectives In this chapter, you will learn:
CSCI 351 – Mobile Applications Development
Mobile Data Solutions Inc
Architecture of Android
Chapter 2: Operating-System Structures
Outline Introduction Related Work
Operating System.
Mobile Application Test Case Automation
Chapter 2: System Structures
Chapter 4: Multithreaded Programming
Google translate app demo
Event Studio Cognos 8 BI.
Auto-tagging of Media using Local Bluetooth Information
NEED OF JAILBREAKING IN IOS PENETRATION TESTING
MIT App Inventor By Hans Yip.
Chapter 1: Introduction
Meng Cao, Xiangqing Sun, Ziyue Chen May 28th, 2014
Android training in Chandigarh. What is ADB ADB stands for Android Debug Bridge. It is a command line tool that is used to communicate with the emulator.
There's an App or Wearable for That!
Chapter 4.
International Symposium on Microarchitecture. New York, NY.
Auto-tagging of Media using Local Bluetooth Information
PredictRemainingTime
and LMAP liaison Document Number: IEEE R0
Chapter 2: System Structures
Use of handheld electronic devices for census data collection (Bhutan)
RM3G: Next Generation Recovery Manager
The Pitfalls and Guidelines for Mobile ML Benchmarking
Summary AutoMap™ is an innovative vehicle tracking and reporting system that ensures the vehicle’s location and status. Reduces Cost Removes all routine.
Characterizing Smartwatch Usage In The Wild
Android Developer Fundamentals V2
“Location Privacy Protection for Smartphone Users”
MPJ: A Java-based Parallel Computing System
Chapter 3: Processes.
Chapter 2: Operating-System Structures
Presentation transcript:

Collaborative Detection of Energy Bugs Adam Oliner, Anand Iyer, and Ion Stoica UC Berkeley Eemil Lagerspetz, and Sasu Tarkoma U Helsinki Carat / Collaborative Detection of Energy Bugs 9/7/2018

Mobile is hot … sometimes, literally Carat / Collaborative Detection of Energy Bugs 9/7/2018

State of the Art Kill ALL your background apps Dim the screen Exchange it at the store Look at CPU utilization Uninstall the app and then reinstall it Restart Carat / Collaborative Detection of Energy Bugs 9/7/2018

Related Work Survey of forum posts re: energy bugs Abhinav Pathak, Y. Charlie Hu, Ming Zhang. Bootstrapping Energy Debugging on Smartphones: A First Look at Energy Bugs in Mobile Devices, HotNets’11 Mobile debugging framework proposal Sharad Agarwal, Ratul Mahajan, Alice Zheng, and Victor Bahl. There’s an app for that, but it doesn’t work. Diagnosing Mobile Applications in the Wild, HotNets’10 CarrierIQ, PowerTutor (Android), etc. Carat / Collaborative Detection of Energy Bugs 9/7/2018

Goals Characterize energy use on smartphones Identify energy bugs in the wild Provide data to help diagnose / fix them Carat / Collaborative Detection of Energy Bugs 9/7/2018

Energy Bug An energy bug is a system behavior that causes unexpectedly heavy use of energy and which is not intrinsic to providing the desired functionality. Carat / Collaborative Detection of Energy Bugs 9/7/2018

Collaborative Detection Community specifies expected behavior Statistical deviation is a bug Actionable Carat / Collaborative Detection of Energy Bugs 9/7/2018

What is Intrinsic? Carat / Collaborative Detection of Energy Bugs 9/7/2018

Is App X buggy? Carat / Collaborative Detection of Energy Bugs 9/7/2018

Carat Carat / Collaborative Detection of Energy Bugs 9/7/2018

Carat Architecture Carat / Collaborative Detection of Energy Bugs 9/7/2018

Sampling We collect... battery level, running apps, trigger OS version, device model, uuid +more on Android (cpu usage, app memory fractions, battery temp, battery voltage, ...) ... when... battery level or status changes location changes the user is using Carat Carat / Collaborative Detection of Energy Bugs 9/7/2018

Challenges Data will be difficult to collect Data will be noisy and incomplete Data will be biased Multiple confounding factors Carat / Collaborative Detection of Energy Bugs 9/7/2018

Privacy Concerns Opt-in EULA prior to installation Nothing personally identifying is collected Sampled data shown within the app Carat / Collaborative Detection of Energy Bugs 9/7/2018

5% Battery Level Granularity Deal with uncertainty using known points (BatteryLevelChanged event) Carat / Collaborative Detection of Energy Bugs 9/7/2018

Analysis Carat / Collaborative Detection of Energy Bugs 9/7/2018

Looks like it does, and it's not even parallel Does it scale? Looks like it does, and it's not even parallel Carat / Collaborative Detection of Energy Bugs 9/7/2018

Found 3 separate bugs injected into Mobile Wikipedia Can it find bugs? Found 3 separate bugs injected into Mobile Wikipedia Carat / Collaborative Detection of Energy Bugs 9/7/2018

Can it find real bugs? Found bugs corroborated by forum posts Kidle on 6 of our clients, forums blame WhisperSync over 3G. Fix by turning on WiFi and save 2h battery life Facebook was a bug for 24 of its 63 users No strong correlations, may be caused by workload iOS 5.0.1 used less energy than 5.0 in our data There is debate on which one is better. 5.1 improves on both in our data. Carat / Collaborative Detection of Energy Bugs 9/7/2018

Can it be accurate? Does sampling match reality? “Do not attempt to calculate battery drainage rate or battery time remaining” - iOS API Docs re: UIDeviceBatteryLevelDidChangeNotification Monsoon Power Monitor Ground truth Carat / Collaborative Detection of Energy Bugs 9/7/2018

Battery indicator agrees with the Power Monitor Carat / Collaborative Detection of Energy Bugs 9/7/2018

Battery drain rates distributions are also similar Carat / Collaborative Detection of Energy Bugs 9/7/2018

TestFlight Beta on iOS Carat / Collaborative Detection of Energy Bugs 9/7/2018

Number of Active Users Grew Quickly -20k samples today from up to 75 active users Carat / Collaborative Detection of Energy Bugs 9/7/2018

Top Devices/Carriers/OS versions Carat / Collaborative Detection of Energy Bugs 9/7/2018

Sampling Triggers Carat / Collaborative Detection of Energy Bugs 9/7/2018

Some Hogs/Bugs Carat / Collaborative Detection of Energy Bugs 9/7/2018

Next Steps Expand beta testing App Store API for developers Android port Platform for collaborative debugging Carat / Collaborative Detection of Energy Bugs 9/7/2018

End of Presentation carat.cs.berkeley.edu github.com/amplab/carat carat@eecs.berkeley.edu eemil.lagerspetz@cs.helsinki.fi Submitted to Mobicom Poster in NSDI HotDep in progress Carat / Collaborative Detection of Energy Bugs 9/7/2018