Kunal Agrawal, PhD, assistant professor of computer science & engineering in the School of Engineering & Applied Science at Washington University in St. Louis, has won a prestigious Faculty Early Career Development Award (CAREER award) from the National Science Foundation (NSF).
The awards are given “in support of the early career-development activities of those teacher-scholars who most effectively integrate research and education within the context of the mission of their organization” with the goal of “building a firm foundation for a lifetime of integrated contributions to research and education.”
According to Ralph S. Quatrano, PhD, dean of the School of Engineering & Applied Science, 12 engineering faculty have received a CAREER award since 2005, including nine of the 20 faculty in the Department of Computer Science & Engineering.
“This truly remarkable achievement is a significant recognition of the tremendous quality of our faculty,” Quatrano says.
The goal of Agrawal’s project, titled “Provably Good Concurrency Platforms for Streaming Applications,” is to design platforms that will allow programmers to easily write correct and efficient high-throughput parallel programs.
In particular, her platforms will be useful for data-intensive applications, such as audio, video and signal processing, allowing these applications to run on modern multicore machines.
Most computers made in the past decade are multicores; that is, they contain multiple processing elements, or cores. In addition, the number of cores on machines is increasing at close to an exponential rate.
Traditional sequential programs cannot make use of more than one core at a time, potentially wasting resources. In order to execute efficiently on these machines, programmers must write parallel programs, programs capable of using more than one core at the same time.
Agrawal’s project concentrates on designing platforms that can efficiently execute a class of parallel programs called streaming programs. In streaming applications, a set (potentially ordered) of operations is applied to each element in a data set (the stream).
Agrawal’s work will enable automatic management of synchronization and scheduling of operations, so that the programmers can write their programs at a high level without worrying about those details. A large class of data intensive applications, including bioinformatics applications, and many scientific applications, as well as audio, visual and image processing, can be expressed as streaming programs.
“This research will advance the state-of-the-art in streaming platforms, both theoretically and practically,” Agrawal says. “While the research is theoretical, we want to design platforms that both academic and industry partners will want to implement in their streaming systems.”
The research will support both graduate and undergraduate research as well as Agrawal’s.
In addition, the research will be integrated into Agrawal’s graduate course “Theory of Parallel Systems.” Agrawal also plans to incorporate parallal algorithms in the undergraduate algorithms and data structures course she teaches. Finally, in her free time, Agrawal designs puzzles that indirectly teach students about parallel computing concepts.
Agrawal joined WUSTL after earning a PhD in computer science from the Massachusetts Institute of Technology. She earned a master’s degree in computer science from the National University of Singapore and a bachelor’s degree from Mumbai University.