Can HP & Intel build the world’s most efficient data center?

As more energy management systems become automated, the search for a data center design that doesn’t gobble up absurd amounts of energy is heating up. 

For those organizations also concerned with serious scientific research, the issue is even more pressing. Big problems demand powerful computers, but these computers need lots of power and the perfect temperature in order to operate at top speeds.

The US Department of Energy’s National Renewable Energy Laboratory recently used its governmental clout to recruit two top IT companies to aid in its quest for a super efficient data center.

 NREL has asked Hewlett-Packard and Intel to provide help it design a new energy-efficient high performance computer (HPC) system dedicated to energy systems integration, renewable energy research, and energy efficiency technologies.

According to NREL, the new HPC system will cost about $10 million and will reside at the Energy Systems Integration Facility (ESIF), currently under construction on the agency’s Golden, Colorado, campus. Every effort was made to keep the data center’s design very compact, thus reducing the distance electrical and plumbing components must run. 

The project features a technology, currently under development, that uses warm water in the computing rack to efficiently cool the servers. Waste heat from the computer system will be used as the primary heat source in the ESIF offices and lab space, and possibly adjacent buildings.

This and other advanced design features mean the new data center could easily be the world’s most energy efficient, with a possible annualized average power usage effectiveness (PUE) rating of 1.06 or better. (The average data center operates with a PUE of 1.91, according to 2009 data from the Environmental Protection Agency’s Energy Star Program.) 

But that’s if they can really achieve it in real life. Some are skeptical that what seems like a powerful triple partnership will fail to deliver on its claims.

One thing’s for certain, however: the HPC’s petascale computing capability (1 million billion calculations per second) is the world’s largest computing capability dedicated solely to renewable energy and energy efficiency research.

* Beth Buczynski, EarthTechling