News

Solution for Big Data


28 October 2014


Australian cloud-computing experts based at the University of Sydney will work with researchers in the United Arab Emirates to find better methods for exploring and exploiting large quantities of stored data.

Professor Albert Zomaya, Chair of High Performance Computing and Networking, and his colleague Dr Young Choon Lee from the School of Information Technologies have been awarded a $920,000 ICT Fund grant to support their work into the design and development of scalable solutions for the storage and accessing of large volumes of data. The pair will work in collaboration with researchers from Khalifa University, UAE, headed by Dr. Paul Yoo.

According to Professor Zomaya 90 percent of the globe 's current data was created over the past few years and with emerging internet technologies and associated devices the upward trend will continue for the foreseeable future.

"Over the past decade, intelligent data models have been widely used in a number of massive and complex data-intensive fields such as astronomy, biology, climatology, medicine, and finance," he says.

However, 60 percent of industry executives report they are not effectively using their company 's stored data.

"Our models will maintain if not improve the predictive and scalable performances of data storage.

The underlying principle of our work is to increase locality and minimise data movement to reduce network traffic," says Professor Zomaya.

To minimise energy consumption the group plans to design models and algorithms with the least processing and memory requirements.

The research team also plans to optimise the processing of big data by integrating intelligent machine-learning research with clouds, giving it more flexibility to rely on sites in the cloud to provide a computation-hosting framework to support virtual machine-hosting services.

Both of these says Professor Zomaya represent basic Cloud infrastructure.

This will allow users to apply their own applications to the data and their resources can be accessed as and when users need them," states Professor Zomaya.

"It is important to provide scalability that will allow businesses to gauge cloud resources up/out or down depending on their enterprise 's need."

This would free the business from procuring resources for peak or future needs as cloud resources would be available in minimal time.

Professor Zomaya is a lecturer in IT Research Methods at the School of Information Technologies.


Follow University of Sydney Media on Twitter

Media enquiries: Victoria Hollick, victoria.hollick@sydney.edu.au, 0401 711 361