Two iphones.
News_

Sharing resources could result in lower mobile costs

20 November 2017
Award-winning paper outlines new mobile cloud computing solution

Sharing your mobile phone's resources with people around you could cut your mobile costs, claim researchers from the University of Sydney's School of Information Technologies.

In collaboration with experts from Zhejiang University, led by Professor Zhaohui Wu and in collaboration with Professors Shuiguang Deng and Jianwei Yin, the researchers have devised a new cloud computing approach that would use the idle resources of one or more mobile devices in a particular area – such as mobile phones, tablet devices and smart watches – to run applications on another mobile device.

Chair Professor of High Performance Computing and Networking Albert Y. Zomaya said the new approach – named RESP (REvenue-driven Service Provision for mobile devices) – has the potential to eliminate mobile devices’ unstable and costly internet connections.

“Traditionally, powerful remote cloud servers are used to accommodate service requests as they have much higher computational ability than individual mobile devices. However, these remote solutions are associated with high overhead costs,” he said.

“Our solution utilises the ubiquitous presence of mobile devices to provide on-demand services anytime, anywhere, and at a fraction of the cost.”

The RESP framework is composed of multiple mobile devices and a trusted broker – such as a telecommunications provider – who serves as an intermediary responsible for communication among the devices.

An illustration of the RESP mobile cloud computing architecture

An illustration of the RESP mobile cloud computing architecture.

As RESP relies on the willingness of mobile device owners to participate, there is a need to employ a proper incentive to motivate people.

“Resource sharing could be encouraged by offering mobile device owners rewards – such as discounts on their monthly phone bill or credit points,” said Dr Wei Li, Research Associate in the School of Information Technologies.

To evaluate the performance of RESP, the researchers have conducted a set of experiments.

“We implemented the algorithm in Python and conducted several experiments. Our experiments demonstrated that our algorithm outperforms standard benchmarks and provides the best overall performance for mobile device users,” said Dr Li.

In the near future, the researchers plan to undertake real-world testing of RESP and fine tune its parameters to improve performance.

A paper outlining the RESP framework won the prestigious Best Paper Award at the International Conference on Service-Oriented Computing (ICSOC 2017), held in Spain last week. This was the only Best Paper Award for this year’s event.

The University of Sydney had six additional papers accepted to ISCOC 2017 – the highest number accepted from any university.

Jennifer Peterson-Ward

Media and PR Adviser (Engineering and IT)

Related articles