Rice wins $3M NSF grant for supercomputer

‘Workhorse’ system will include Rice’s first 3-D visualization facility

Rice News staff

A wide-ranging team of 53 Rice faculty have won almost $3 million from the National Science Foundation (NSF) for a powerful new high-performance computer system that will feature Rice’s first 3-D visualization studio.

The new computer is dubbed DAVinCI, which stands for Data Analysis and Visualization Cyber-Infrastructure for Computational Science and Engineering Applications. The computer system will be housed in Rice’s state-of-the-art $16 million data center, and grant investigators say they hope the system will be online this fall. The visualization center will be on campus and will be available later in the year.

“This system was designed to be a workhorse platform for scientific and engineering research,” said DAVinCI program principal investigator Alan Levander, the Carey Croneis Professor of Earth Science and chair of the Department of Earth Science. “It will handle two kinds of computational research. The first requires lots of memory and lots of simultaneous parallel processing, and the second requires millions of serial calculations that have to be repeated over and over again.”

The faculty investigators who use DAVinCI come from every department in the schools of Natural Science and Engineering as well as a few departments in Social Sciences. Although their research runs the gamut from the nano to the macro and from the theoretical to the applied, Levander said a unifying theme of the research is a need for data analysis, data interpretation and data visualization.

“To support the data-intensive needs of our faculty, we’re going in a new direction,” said Jan Odegard, executive director of Rice’s Ken Kennedy Institute for Information Technology. “DAVinCI is more than just another large computational resource here on campus. It will be designed and built to help researchers make sense of vast amounts of observational, experimental and theoretical data.”

Officials said the new system will be useful for researchers involved with collaborative research in the Texas Medical Center.

“Rice has a strong set of supercomputing resources, and the addition of the DAVinCI system will allow us to build upon our partnerships in the Texas Medical Center and upon our initial success in the BioScience Research Collaborative,” said Mary “Cindy” Farach-Carson, associate vice provost for research and professor of biochemistry and cell biology. “In particular, the ability to visualize reconstructions of rich data sets will provide our faculty with a unique resource for the impending reality of personalized medicine.”

Odegard, who played a key role in developing the concept and proposal for DAVinCI, said the system is “the natural next step” for Rice’s vision of supporting shared research computing infrastructure.

“We are building on the strength of Rice’s faculty, and leveraging the partnership between the Ken Kennedy Institute and the division for Information Technology,” Odegard said.

Kamran Khan, Rice’s vice provost for information technology, said DAVinCI represents another important piece of Rice’s groundbreaking efforts in research computing.

“The partnerships we have formed at Rice allow us to deliver critical computational resources in support of cutting-edge research,” Khan said.

DAVinCI’s other co-principal investigators are Danny Sorensen, the Noah Harding Professor of Computational and Applied Mathematics; Paul Padley, associate professor of physics and astronomy; Jamie Padgett, assistant professor in civil and environmental engineering; and Oleg Igoshin, assistant professor in bioengineering.

The visualization portion of DAVinCI will display 3-D images in a dedicated visualization lab. Researchers will be able walk through the lab, view the images from different angles and perhaps even walk into the image and tag particular portions for later study.

Levander said the visualization technology, which has been used in industry for years, can let investigators see correlations they wouldn’t otherwise see.

“There are probably dozens of things that people have in mind already, but I would be willing to bet that we don’t yet know what the most exciting applications of the visualization technology will be,” he said.

Khan and Odegard said the goal is to deploy a system that can meet today’s needs as well as the challenges of the future.

“With NSF’s support, Rice is meeting the challenges of data-intensive computing in a cost-effective way that has maximum benefit for the most faculty possible,” Odegard said.

The NSF funding was made possible by the American Recovery and Reinvestment Act.


About Jade Boyd

Jade Boyd is science editor and associate director of news and media relations in Rice University's Office of Public Affairs.