Michelle Goodstein

I recently defended my PhD in Computer Science, as a member of the Log-Based Architecture project, advised by Todd Mowry. My research interests involve developing new theoretically sound frameworks to tackle important systems problems, with a particular emphasis on parallel processing and sophisticated code analysis. My thesis research focused specifically on the problem of how to create a powerful yet efficient dynamic code analysis framework that can find software bugs in parallel programs as they execute.

My main research contribution is dataflow analysis-based parallel monitoring, an entirely new approach to dynamic application monitoring with advantages similar to dataflow analysis, but in the entirely different context of online parallel monitoring. Using insights from dataflow analysis, my frameworks enable parallel application monitoring without capturing a total order of application instructions across parallel threads.

I am currently a Research Scientist at Facebook.

Publications:

Tracking and Reducing Uncertainty in Dataflow Analysis-Based Dynamic Parallel Monitoring
Michelle L. Goodstein, Phillip B. Gibbons, Michael A. Kozuch and Todd C. Mowry
To appear in International Conference on Parallel Architectures and Compilation Techniques (PACT 2015)

Chrysalis Analysis: Incorporating Synchronization Arcs in Dataflow-Analysis-Based Parallel Monitoring
Michelle L. Goodstein, Shimin Chen, Phillip B. Gibbons, Michael A. Kozuch and Todd C. Mowry
In International Conference on Parallel Architectures and Compilation Techniques (PACT 2012) [pdf] [ slides ]

Simulating Multi-Million-Robot Ensembles
Michael P. Ashley-Rollman, Padmanabhan Pillai and Michelle L. Goodstein
In International Conference on Robotics and Automation (ICRA 2011) [pdf]

Butterfly Analysis: Adapting Dataflow Analysis to Dynamic Parallel Monitoring
Michelle L. Goodstein, Evangelos Vlachos, Shimin Chen, Phillip B. Gibbons, Michael Kozuch and Todd C. Mowry.
In International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2010) [pdf] [slides]

ParaLog: Enabling and Accelerating Online Parallel Monitoring of Multithreaded Applications
Evangelos Vlachos, Michelle L. Goodstein, Michael Kozuch, Shimin Chen, Babak Falsafi, Phillip B. Gibbons and Todd C. Mowry.
In International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2010) [pdf]

CMU-CS-07-134
A Two Player Game to Combat Web Spam
Michelle Goodstein, Virginia Vassilevska
[ps, pdf]
Please see copyright notice .

Courses:

Machine Learning (Fall 2005)
Algorithms in the Real World (Fall 2005)
Graduate Algorithms (Spring 2006)
Introduction to Theoretical Cryptography(Spring 2006)
Computational Complexity Theory(Fall 2006)
Computational Molecular Biology and Genomics(Fall 2006)
Graduate Computer Networking(Spring 2007)
Computer Architecture(Fall 2007)
Semantics of Programming Languages(Spring 2008)
Program Analysis (Spring 2010, audit)
Optimizing Compilers for Modern Architectures (Spring 2011, audit)

Teaching:

TA for Graduate Architecture (Fall 2009)
TA for Undergraduate Algorithms (Spring 2007)