Group Leads Workshop in Philippines While Using Supercomputer at ODU
September 21, 2018
Workshop participants learn high-end data analysis using ODU's Turing supercomputer
At a recent workshop led by Old Dominion faculty and students, participants learned about genetic analysis using the University's supercomputers, all while in the Philippines.
The two-week "Omics and Bioinformatics" was part of a $4.6 million, five-year Partnership for International Research and Education (PIRE) award won by Kent Carpenter, a professor and marine biologist in biological sciences.
Established in 2005 by the National Science Foundation, PIRE was created to support international research and education and encourage a working-class STEM workforce.
Carpenter is studying human impacts on genetic and species diversity of fish in the Philippines over the last century.
He led the workshop along with Dan Barshis, assistant professor of biological sciences; Dave Gauthier, associate professor of biological sciences; senior Ivan Lopez; and recent graduate Amanda Ackiss.
Workshops help U.S.-based researchers share what they know, but analyzing genetic data requires fast computers with large memories. These are hard to come by in developing countries like the Philippines.
In earlier seminars, instructors used computers that were available locally and made adjustments.
"Oftentimes in science, you can trim the data to make it fit on a computer," said Wirawan Purwanto, a computational scientist in information technology services at ODU. "But if you use a big computer, the machine can handle the real data. Things can look quite different when looking at reduced data rather than real size."
During the most recent workshop more than 30 Philippine graduate students, research assistants, faculty and government researchers logged in remotely to Turing, the ODU supercomputer located on campus.
Like a web browser, participant's individual computers acted as screens 9,000 miles away from the processors and servers in Norfolk.
With processing speeds more than eight times faster than a desktop computer, Turing provided high-end, fast data analysis with a simple login. Students could see and learn much more.
"Aside from some software uploads it was really seamless," Purwanto said. "Before they had to skip steps and change the data to make it work for the computers they had. Now the students are getting a real-world experience of working with genetic data."
It wasn't only the students who benefited. According to Barshis, Turing changed the experience for instructors, too.
"It would not have been possible to do all our analyses using a locally installed system, so using Turing was vastly preferable from a teaching standpoint. Everything worked really, really well," he said.
