An exploratory study on assessing the energy impact of logging on Android applications

An exploratory study on assessing the energy impact of logging on Android applications Execution logs are debug statements that developers insert into their code. Execution logs are used widely to monitor and diagnose the health of software applications. However, logging comes with costs, as it uses computing resources and can have an impact on an application’s performance. Compared with desktop applications, one additional critical computing resource for mobile applications is battery power. Mobile application developers want to deploy energy efficient applications to end users while still maintaining the ability to monitor. Unfortunately, there is no previous work that study the energy impact of logging within mobile applications. This exploratory study investigates the energy cost of logging in Android applications using GreenMiner, an automated energy test-bed for mobile applications. Around 1000 versions from 24 Android applications (e.g., Calculator, FeedEx, Firefox, and VLC) were tested with logging enabled and disabled. To further investigate the energy impacting factors for logging, controlled experiments on a synthetic application were performed. Each test was conducted multiple times to ensure rigorous measurement. Our study found that although there is little to no energy impact when logging is enabled for most versions of the studied applications, about 79% (19/24) of the studied applications have at least one version that exhibit medium to large effect sizes in energy consumption when enabling and disabling logging. To further assess the energy impact of logging, we have conducted a controlled experiment with a synthetic application. We found that the rate of logging and the number of disk flushes are significant factors of energy consumption attributable to logging. Finally, we have examined the relation between the generated OS level execution logs and mobile energy consumption. In addition to the common cross-application log events relevant to garbage collection and graphics systems, some mobile applications also have workload-specific log events that are highly correlated with energy consumption. The regression models built with common log events show mixed performance. Mobile application developers do not need to worry about conservative logging (e.g., logs generated at rates of ≤ 1 message per second), as they are not likely to impact energy consumption. Logging has a negligible effect on energy consumption for most of the mobile applications tested. Although logs have been used effectively to diagnose and debug functional problems, it is still an open problem on how to leverage software instrumentation to debug energy problems. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Empirical Software Engineering Springer Journals

An exploratory study on assessing the energy impact of logging on Android applications

Loading next page...
 
/lp/springer_journal/an-exploratory-study-on-assessing-the-energy-impact-of-logging-on-IpqWO6h3Si
Publisher
Springer Journals
Copyright
Copyright © 2017 by Springer Science+Business Media, LLC
Subject
Computer Science; Software Engineering/Programming and Operating Systems; Programming Languages, Compilers, Interpreters
ISSN
1382-3256
eISSN
1573-7616
D.O.I.
10.1007/s10664-017-9545-x
Publisher site
See Article on Publisher Site

Abstract

Execution logs are debug statements that developers insert into their code. Execution logs are used widely to monitor and diagnose the health of software applications. However, logging comes with costs, as it uses computing resources and can have an impact on an application’s performance. Compared with desktop applications, one additional critical computing resource for mobile applications is battery power. Mobile application developers want to deploy energy efficient applications to end users while still maintaining the ability to monitor. Unfortunately, there is no previous work that study the energy impact of logging within mobile applications. This exploratory study investigates the energy cost of logging in Android applications using GreenMiner, an automated energy test-bed for mobile applications. Around 1000 versions from 24 Android applications (e.g., Calculator, FeedEx, Firefox, and VLC) were tested with logging enabled and disabled. To further investigate the energy impacting factors for logging, controlled experiments on a synthetic application were performed. Each test was conducted multiple times to ensure rigorous measurement. Our study found that although there is little to no energy impact when logging is enabled for most versions of the studied applications, about 79% (19/24) of the studied applications have at least one version that exhibit medium to large effect sizes in energy consumption when enabling and disabling logging. To further assess the energy impact of logging, we have conducted a controlled experiment with a synthetic application. We found that the rate of logging and the number of disk flushes are significant factors of energy consumption attributable to logging. Finally, we have examined the relation between the generated OS level execution logs and mobile energy consumption. In addition to the common cross-application log events relevant to garbage collection and graphics systems, some mobile applications also have workload-specific log events that are highly correlated with energy consumption. The regression models built with common log events show mixed performance. Mobile application developers do not need to worry about conservative logging (e.g., logs generated at rates of ≤ 1 message per second), as they are not likely to impact energy consumption. Logging has a negligible effect on energy consumption for most of the mobile applications tested. Although logs have been used effectively to diagnose and debug functional problems, it is still an open problem on how to leverage software instrumentation to debug energy problems.

Journal

Empirical Software EngineeringSpringer Journals

Published: Sep 21, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off