Green supercomputer pioneer to model micro drones

US Air Force backs Wu Feng for fluid dynamics modelling

By Richard Chirgwin, 7th February 2013

Cloud storage: Lower cost and increase uptime

The US Air Force Office of Scientific Research has slung some money in the direction of Virginia Tech to conduct fluid dynamics modelling as part of its Micro Air Vehicle – to you and I, micro-drones – projects.

The wrinkle in the contract, worth up to $US6 million over five years, is that researcher Wu Feng is to bring his skills in low-power supercomputing to the research.

Feng’s CV includes the development of the Green 500 supercomputer list, the creation in 2002 of the Green Density cluster, and more recently, HokieSpeed. At number 11 on the worldwide 2011 Green 500 list, the 2,500-core, 185,000-GPU core HokieSpeed was at the time the most energy-efficient super in America, and was put together at a cost of just $US1.4 million.

Back to the current project: the USAF wants better simulation speed for the fluid dynamics research necessary to develop its MAVs, and believes Feng’s work with HokieSpeed means he fits the bill.

Feng believes there’s still a lot of work to be done to extract the most out of parallel hardware and software that exploits it. “Furthermore, coupling hardware-software co-design with advances in algorithmic innovation offers the promise of multiplicative speed-ups,” he says in this press release.

As well as computer scientists, Feng is assembling a team that includes mechanical engineers, aerospace engineers, and mathematicians. ®

Customer Success Testimonial: Recovery is Everything