Dynamic Concurrency Control

Spring 2008

Student
Alex Chao
Advisor
Bill Scherlis
Project description

With the increase in multi-processor systems today, the need for accurate, foolproof concurrent programs is a necessity. Even with the most precise and meticulous programming, potential problems in concurrency such as race conditions are still extremely hard to spot. Many of these overlooked problems end up costing companies millions of dollars. New analysis-based concurrency control software such as FLUID, led by Professor Bill Scherlis, has emerged to help combat this issue. My focus for my research will be to understand fully how programs like FLUID operate and look into the next step---dynamic concurrency control, which will provide an even more accurate testing mechanism for concurrent programs. With the help of resources such as the book Java Concurrency in Practice by Brian Goetz, I will document my research through reports as well as presentations. I will also be expected to write sample Java code that demonstrates both safe and unsafe concurrency programming practices.

Return to project list