Welcome Your IET account
AppStreamer illustration

Mobile apps could be streamed ‘instantly’ to phones

Image credit: Jamayal Tanweer

University of Purdue engineers have developed software, called 'AppStreamer', which allows for data-heavy apps to be accessed on a smartphone without using up internal storage.

The software is based on the idea that most apps do not need all their files at any one time, so just a small fraction of the files actually need to be stored on the device itself. The software continuously predicts when it is necessary for an app to access resources from a cloud server when the app is used, allowing the app to use only the space it needs on a phone’s internal storage at any given time.

“It’s like how Netflix movies aren’t actually stored on a computer. They are streamed to you as you are watching them,” said Professor Saurabh Bagchi, director of Purdue University’s Center for Resilient Infrastructures, Systems and Processes. “Here, the application components, like heavy video or graphics or code paths, are streaming instantly, despite the errors and slowdowns that are possible on a cellular network.”

Bagchi and his colleagues tested the software using gaming apps, which tend to be relatively heavy.

They used AppStreamer to run two popular Android games: Dead Effect 2 and Fire Emblem Heroes. They found that the software cut down storage requirements by 87 per cent and 86 per cent respectively. Importantly, the software was able to shuffle data between the cloud server and app without stalling the games, with a large majority of study participants not noticing any difference between playing with all files stored on-device vs. playing with AppStreamer.

According to Bagchi, given that the software works so seamlessly for these gaming apps, it could work just as well for other apps which typically require less storage.

AppStreamer could prove to be a valuable tool in the 5G age. For instance, it could allow for data-heavy gaming apps to be downloaded in an instant, taking up minimal space on a device. Bagchi and his colleagues also suggest that the software could prove valuable in the Internet of Things space, for example by helping autonomous vehicles pull data from servers in milliseconds to respond to their surroundings while in motion.

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles

Info Message

We use cookies to give you the best online experience. Please let us know if you agree to all of these cookies.


Learn more about IET cookies and how to control them