No. 1 Story

ACCC clears Optus to scrap HFC network and use NBN instead

The ACCC has cleared, provisionally, the proposed deal between Optus and NBN Co under which Optus is to be paid around $800m to shut down its HFC network and transfer customers onto the NBN. read more

Research promises to slash data centre energy consumption

IT Industry - Market

University of Sydney researchers have found a way of simultaneously optimising the energy consumption and completion time of computationally intensive processes.

Dr Young Choon Lee and Professor Albert Zomaya of the University of Sydney's Centre for Distributed and High Performance Computing have patented a scheme that is said to reduce CPU power consumption by as much as 60%.

Lee and Zomaya's patented Energy Conscious Scheduling algorithm (ECS) is aimed at computationally intensive tasks such as those performed in areas such as astronomy, chemistry, life sciences and financial forecasting.

The basic idea appears to be to take advantage of the dynamic voltage scaling (DVS) feature of modern processors. Reducing the voltage supply and clock frequency means a process takes longer to complete but uses less energy.

ECS uses this to adjust the speed of various processes to reduce the need for one to wait for another to complete.

"Computations are typically comprised of interdependent tasks, so the need to wait for a parent task to complete can create slack and therefore wastage," said Zomaya.

"When ECS is employed with the help of DVS capability, mapping decisions between processors, supply voltages, and tasks are streamlined to significantly lower the amount of energy required at any given time," he added.

Reducing the slack time (when processors are not usefully employed) in this way can cut energy consumption by between 10 and 60%, Zomaya claimed.

Lee and Zomaya are developing an ECS prototype, which they hope to commercialise by 2011.

"ECS will work well on cloud systems as they mature but will give a more immediate means for organisations to reduce their data centre's energy consumption. With demand for computing and data storage capacities growing exponentially, this world first will make a valuable contribution to both the environment to and organisations' bottom line," said Zomaya.

Power consumption by servers has more than doubled since 2000, according to University of Sydney officials.

The availability and price of electricity has become a significant factor in choosing sites for large data centres. In addition to cost, there is growing social pressure for organisations to reduce their energy consumption or at least make more efficient use of what they use.