November 9, 2018 10:09 am
Updated: November 9, 2018 10:17 am

University of Alberta research could prevent app updates from draining smartphone batteries

WATCH ABOVE: A pair of University of Alberta studies embarked on finding ways to reduce how much certain apps drain your phone's battery. Slav Kornik has the details.


New University of Alberta studies suggest more can be done to ensure app updates don’t drain your phone’s battery as quickly and don’t use up as much memory.

Department of computing science assistant professor Abram Hindle and his team examined Android apps to determine which ones are making use of machine learning algorithms and how the algorithms can cause extra drain on your battery.

Story continues below

Hindle said about five per cent of popular apps engage in machine learning, which is a type of artificial intelligence that learns as it identifies new patterns in data, without being programmed.

“As a software engineer, there’s a design tradeoff: what do you want to pay in terms of energy for the computation that you do?” Hindle said.

“So you basically say, ‘I want this much accuracy and thus I will have to pay around this much energy. Whereas if I want more accuracy, I’m going to have to pay more energy.'”

WATCH: Why your smartphone battery keeps dying

Smartphone apps use machine learning algorithms for tasks such as predictive typing, spam email detection and facial recognition on your camera. They can make apps more intuitive and reactive, but they can also eat up your phone’s battery power and memory.

Hindle’s research found machine learning algorithms result in more responsive apps, but also increased energy consumption and decreased battery life.

“There are many reasons why an end user would care about the energy efficiency of machine learning algorithms on a phone,” Hindle said. “The main outcome is, the more accurate we want things to be, the more energy they’ll use. You end up paying for accuracy.”

READ MORE: How to take care of your smartphone’s battery in the summer heat

Computing science graduate student Shaiful Chowdhury has developed a new tool, designed to better predict and improve software’s effects on battery life.

Chowdhury said GreenScaler is able to simulate the energy consumption of software, without hooking them up to a device and measuring the usage while in operation.

“There are reports that say, users mostly care about software energy efficiency now. They don’t want to install apps that drain your battery faster. So now developers need to optimize the energy efficiency of their apps,” he said.

READ MORE: iPhone update will let you turn off battery slowdowns; experts suggest you think twice

Chowdhury said his tool could allow small developers to gauge how their software will affect the battery life of users’ phones, and optimize their pass to reduce the impact of their battery.

“While companies like Facebook likely have the resources to measure energy consumption on physical devices, for smaller companies, measuring software energy consumption using current tools is extremely expensive and difficult,” said Chowdhury.

Chowdhury and Hindle hope GreenScaler will result in developers being able to optimize their software to allow users to use their phones longer before needing to recharge their batteries.

© 2018 Global News, a division of Corus Entertainment Inc.

Report an error


Want to discuss? Please read our Commenting Policy first.