|
|
| Ph.D. of Computer Science, Rutgers University (May 2007) | |
| Thesis: Just-in-time and Just-in-place Deadlock Resolution | Thesis Committee(2007): Michael Littman(Chair), Louis Steinberg, Marvin Paull, Gertrude Levine |
| Qualifier: Exploiting Runtime Exceptions and Static Analyses to Detect Deadlock in Multithreaded Java Programs | Qualifier Committee(2002): Barbara Ryder(Chair), Alex Borgida, Richard Martin, Eric Allender |
The fun of research, as far as I am concerned, is to appreciate common phenomena from a different perspective. Because deadlocks are an established research area appearing in textbooks, many people consider them a solved problem. However, there are still quite a few open research issues in deadlocks. So, my recent research is focused on exploiting various effective techniques such as language design and machine learning to detect, represent, and resolve deadlocks.
As I see it, it is necessary to make one's research work not only interesting but also useful. Java is a popular object-oriented programming language, and most of my recent research work applies to the resolution of Java deadlocks, which are not uncommon even in production systems. It is also important to generalize one's research findings. Deadlocks are a liveness problem: deadlocked processes cannot make forward progress. In addition, many safety properties can be reduced to deadlock freedom.
Please send me an email---try to click somewhere on this webpage to find my email address---if you are interested in my research.