News

Clouds and green lining


8 February 2013

Professor Albert Zomaya: "Global warming and climate change trends call for urgent action to manage information and communication technologies in a sustainable manner."
Professor Albert Zomaya: "Global warming and climate change trends call for urgent action to manage information and communication technologies in a sustainable manner."

Newly-developed cloud computing algorithms may be the solution to reducing the increasing energy consumption and cost of high performance computing networks according to Professor Albert Zomaya, IT expert at the University of Sydney.

Professor Zomaya, director of High Performance Computing and Networking at the University's Faculty of Engineering and Information Technologies, will address this week's Australian Data Centre Strategy Summit and argue the need to move to more ecologically friendly practices of computing, such as cloud computing.

"Before we design and build computing systems, we should have sustainable models for digital ecosystems front and centre of our minds," says Professor Zomaya.

"Green distributed computing involves a number of very complex issues that will be around for some time.

"Global warming and climate change trends call for urgent action to manage information and communication technologies in a sustainable manner by minimising energy consumption and utilising resources more efficiently."

In his address Professor Zomaya will propose solutions to reduce energy consumption such as server consolidation using virtualisation and cloud computing.

"Cloud computing is converging technology. It's on-demand computing, self-organising. It is perhaps the most efficient method of using our computing energy resources," argues Professor Zomaya.

Distributed computing environments such as clusters, grids, clouds, have become the de facto platforms for many applications.

These systems bring a range of heterogeneous resources that should be able to function continuously and autonomously, but often come at a cost: distributed systems can expend a lot of energy. Some HPC systems can use megawatts of electricity for their operation and cooling. On average power bills for such systems run in the millions per year, says Professor Zomaya.

"At the moment we are concerned about the costs of energy bills, system acquisition and maintenance. We want reliability and scalability, but simultaneously we need to be investigating how to reduce the carbon emission rate and e-waste of our complex networks," he says.

Professor Zomaya will reveal the development of new algorithms and tools for energy-aware resource management allocation for large-scale distributed systems, such as clouds, enabling these systems to become environmentally friendly.


Follow University of Sydney Media on Twitter

Media enquiries: Victoria Hollick, 9351 2579, 0401 711 361 victoria.hollick@sydney.edu.au