SeminarsView all Seminars | Download ICal for this event
Learning-Based Controlled Concurrency Testing
Speaker: Suvam Mukherjee
Date/Time: Jun 01 11:00:00
Faculty Advisor: Deepak D'Souza
Concurrency bugs are notoriously hard to detect and reproduce. Controlled concurrency testing (CCT) techniques aim to offer a solution, where a scheduler explores the space of possible interleavings of a concurrent program looking for bugs. Since the set of possible interleavings is typically very large, these schedulers employ heuristics that prioritize the search to "interesting" subspaces. However, current heuristics are typically tuned to specific bug patterns, which limits their effectiveness in practice.
In this work, we present QL, a learning-based CCT framework where the likelihood of an action being selected by the scheduler is influenced by earlier explorations. We leverage the classical Q-learning algorithm to explore the space of possible interleavings, allowing the exploration to adapt to the program under test, unlike previous techniques. We have implemented and evaluated QL on a set of microbenchmarks, complex protocols, as well as production cloud services. In our experiments, we found QL to consistently outperform the state-of-the-art in CCT.
This is joint work with Pantazis Deligiannis (Microsoft Research), Arpita Biswas (Indian Institute of Science) and Akash Lal (Microsoft Research).
Microsoft Teams Meeting Link:
Suvam Mukherjee is a postdoctoral researcher at Microsoft Research India (MSRI). His research interests lie in the areas of programming languages, verification and model-checking. At MSRI, he has been working on several projects in Microsoft Coyote, an open-source framework used by several teams in Microsoft Azure to write reliable and efficient cloud services. Prior to joining MSRI, Suvam obtained his PhD from the Indian Institute of Science, where his thesis focused on developing efficient static analyses for multithreaded programs. For a part of his work, he won the Radhia Cousot Young Researcher Best Paper Award at the 24th Static Analysis Symposium 2017, New York City.
Host Faculty: Deepak D'Souza