HokieSpeed: How GPUs and CPUs Team Up in Supercomputers

Wu Feng, associate professor of computer science and electrical engineering at Virginia Tech, gives a video tour of the VT supercomputer, HokieSpeed. The supercomputer launched late last year, is up to 22 times faster and yet a quarter of the size of the previous supercomputer at VT, titled “X.” Feng says the new supercomputer was designed to be faster, more energy efficient and use much less space. HokieSpeed contains 209 nodes, or separate computers, connected to one another in and across large metal racks, each roughly 6.5 feet tall, to create a single supercomputer that occupies half a row of racks in a large university computer machine room. X took three times the rack space. HokieSpeed’s performance has a single-precision peak of 455 Teraflops, or 455 trillion operations per second, and a double-precision peak of 240 Teraflops, or 240 trillion operations per second. The video runs 3:48.

For additional video, check out our DCK video archive and the Data Center Videos channel on YouTube.

RELATED POSTS:

• NVIDIA Unveils GPUs to Power Clouds
• Submerged Supermicro Servers Accelerated by GPUs
• Jaguar Transforms Into Titan With Nvidia GPUs
• Supercomputers Extend Frontiers of Research
• Greening Supercomputers

About Colleen Miller

Colleen Miller is a journalist and social media specialist. She has more than 25 years of writing and editing experience, with her most recent work dedicated to the online space. Colleen covers the data center industry, including topics such as modular, cloud and storage/big data.