Big-Data Project Nets Professor a Big Award

Professor Florin Rusu’s passion for analyzing voluminous amounts of data has won him a prestigious early-career award from the federal Department of Energy, making him the fifth of the nine eligible faculty members in the computer science and engineering group to earn the honor, and the first one to get this award from the DoE.

Rusu, with the School of Engineering, receives $750,000 over the next five years for his project “Scalable and Energy-Efficient Methods for Interactive Exploration of Scientific Data,” in which he will research novel methods and algorithms to be integrated in a big-data system for scientific analysis.

Interactive exploration of data from experiments or simulations is difficult because the size of the data makes executing even a single query extremely time consuming, the structures of data vary across different scientific disciplines, and analysis algorithms are complex and continuously evolving, Rusu explained.

“The Department of Energy has many very interesting projects and generates massive amounts of data,” Rusu said, pointing to astronomical observations as just one area where he might work. “They have observatories taking high-resolution pictures of the entire sky. Every pixel contains many megabytes of data, but how do you extract the useful information out of it?”

School of Engineering Dean Daniel Hirleman expressed his pride in Rusu’s accomplishment.

“This is just more evidence of our faculty members’ excellence,” Hirleman said. “The campus has won 17 early-career awards so far, with seven of those in the School of Engineering. The concentration in computer science is certainly among the highest percentage of any program nationwide.  Our computer science program is strong and growing, and this is one reason why.”

Only 35, or about 5 percent, of the more than 700 award applications were successful, and only two universities received awards for work in computer science. Other awards went to other junior faculty members at schools including New York University, Rice, UCLA, Duke, Princeton, UC Irvine and Northwestern, as well as to national laboratories.

Rusu has worked on databases most of his career, but now that there is so much more capacity for storing huge amounts of information, big-data analysis has become a fast-growing field that’s in high demand.

“We can store so much more, but then what do we do with it?” he said. “For example, the Large Hadron Collider (at the European Organization for Nuclear Research in Geneva, Switzerland), generates terabytes of information every day. But what are we looking for? We might have hypotheses, but in my field we don’t know precisely what we are looking for until we find it.”

Rusu’s project is designed to devise the tools to make possible the inspection of as many hypotheses as scientists can come up with.

He became interested in his project through astronomy, having heard about a new observatory being designed in South America. It’s scheduled to open in the next few years, but people are already working on the data-management system.

Rusu said he is fortunate to have won this award.

“It feels like a huge accomplishment,” he said. “It allows me to build the tools to do the research I think is needed. The award is a confirmation of my previous work on which this project is built.”