The problem: Increasing amounts of data are being generated that require nearly instantaneous computer processing and feedback. For example, drivers checking road conditions for the morning commute, air traffic controllers tracking the locations of thousands of planes and investors conducting high-frequency trades. When real-time computing fails, it can compromise safety or lead to financial loss.
The researcher: Kyoung-Don Kang, associate professor of computer science at Binghamton University.
The research: Kang is working to make fast-paced data processing more efficient, with help from a National Science Foundation grant of nearly $250,000.
The strategy: “Real-time data is very dynamic and unpredictable,” Kang says. For example, a traffic-monitoring system might not see much activity at midnight Sunday, but it will generate tremendous amounts of data during Monday’s rush hour. That data could slow down the processing system, right when it’s needed most.
Designing systems capable of processing massive amounts of data all the time is not the answer because when the system is idle, resources are wasted. Kang wants to use real-time data applications to sort what must be processed immediately and what can be dropped. “Some data are more important than others,” he says.
Kang will be developing algorithms and software solutions that process the most vital data stream first, prioritizing some operations over others to build more efficient load-shedding and continuous-query processing techniques. This approach can be applied to detect important events, such as unusual traffic patterns or homeland security issues, in real time.