Australian cloud computing specialist Professor Albert Zomaya from the University of Sydney will work with one of China's leading tertiary institution to create more environmentally sustainable data centres.
With some data centres now as large as small suburbs the ongoing energy and environmental impact of the centres prompted the research collaboration between Professor Zomaya and researchers at Tianjin University.
A considerable portion of energy drawn by data centres is being wasted on powering and supporting idle servers
Professor Zomaya who has been awarded an honorary professorship at the University will work with researchers at the Key Laboratory of Smart Grid of Ministry of Education on sustainable computing in Renewable Energy Generation.
“With centres such as the Range International Information Group’s data centre in Langfang, China reportedly covering an area of 6,300,000 square feet it is vital that we address sustainability issues.
“We are confronted with combined challenges of climate change and sustainable development. Our project focuses on how to combine distributed computing with large scale power grids to improve energy efficiency,” says Professor Zomaya.
“With more virtualisation and high-density computing, the energy efficiency of a data centre has become a critical consideration. A big carbon footprint has been created and we aim to minimise it.”
“We are particularly interested in addressing sustainability in different computing and information processing environments and technologies; and at different levels of the computational process.
Professor Zomaya says 52 per cent or more of the power coming into a data centre facility is used directly by the ICT equipment. Server utilisation rates are typically very low, with even idle servers still drawing 60 per cent of its peak performance power.
“A considerable portion of energy drawn by data centres is being wasted on powering and supporting idle servers,” he said.
Potentially there are two energy and cost savings two solutions – either to divide large cloud centre into a number of small data centres or use more intelligent energy resource protocols.
“Our goal is to build ‘green clouds’ or ‘green data centres’, but the computation will be intensive.”
He says the team will employ sophisticated algorithms to maximize the use of clean sources for electricity such as wind-powered or water-powered electricity generation. In this way, the reliance on pollution-generating sources such as coal can be steadily reduced.
School of Electrical Engineering and Automation, Tianjin University is noted for research on smart grids, power distribution networks and information processing. The research outcomes have been widely applied to more than 20 national-level micro-grid projects and the power distribution networks in more than 200 cities.