Piramanayagam Arumuga Nainar 6352 CS&S, 1210 W Dayton St Madison WI 53706 arumuga@cs.wisc.edu http://pages.cs.wisc.edu/~arumuga ______________________________________________________________________________ Research Interests Programming Languages, Software Engineering. Dynamic and Statistical Debugging Methods. ______________________________________________________________________________ Education PhD in Computer Science, expected in May 2010 M.S. in Computer Science, May 2007; GPA: 3.94 Advisor: Prof. Ben Liblit Minor in Mathematics University of Wisconsin-Madison B.E. in Computer Science, April 2005; GPA: 3.7 College of Engineering, Guindy, Anna University, Chennai, India ______________________________________________________________________________ Publications P. Arumuga Nainar, T. J. Rosin and B. Liblit. Statistical Debugging Using Compound Boolean Predicates. In the 2007 International Symposium on Software Testing and Analysis (ISSTA). M. Musuvathi, S. Qadeer, T. Ball, G. Basler, P. Arumuga Nainar, I. Neamtiu. Finding and Reproducing Heisenbugs in Concurrent Programs. In the Eighth Symposium on Operating System Design and Implemenatation (OSDI) 2008. ______________________________________________________________________________ Talks * Statistical Debugging using Compound Boolean Predicates, ISSTA '07 * Automated Adaptive Bug Isolation Using Dyninst - Departmental Summer RA talks, October 2006 - Revised version presented at Paradyn/Condor Week, April 2007 ______________________________________________________________________________ Research Summary Cooperative Bug Isolation for x86 binaries Bug isolation in binary code is necessary for reasons such as the need to debug libraries or non-availability of source code. Performance overhead, which is a critical issue, is reduced using two methods: control flow analysis to find a minimal set of instrumentation points, and self-removing instrumentation. Adaptive Bug Isolation We further reduce performance overhead in binary instrumentation using adaptive instrumentation updates after deployment. At any stage, only a small subset of behaviors is instrumented, resulting in low-overhead monitoring. Propositional Combination of Bug Predicates We augment CBI's bug predictive ability by combining predicates on program state using logical operations (conjunction and disjunction). ______________________________________________________________________________ Work Experience * Research Assistant in the Cooperative Bug Isolation (CBI) project - since Spring 2006 * Research Intern at Microsoft Research - Summer 2007. - Extension of the CHESS model checker to check Windows User Mode drivers. * Teaching Assistant at UW Madison - Fall 2005. - Assistance for Introduction to Data Structures. * Summer Intern at Microsoft India - Summer 2004. - Regression tests for PEAP authentication in the `Routing and Remote Access Server' ______________________________________________________________________________ Honors and Awards UW CS Departmental Summer Graduate Research Assistantship, Summer 2006 All India Rank 2 in GATE 2005 for entrance to graduate studies in India 4th place in the ACM ICPC NCNA Regional Contest, 2006 6th place in the ACM ICPC IIT Bombay Regional Contest, 2005 National Talent Scholarship, also known as NTSE, awarded by NCERT, India Top 0.1% percentile in Mathematics in AISSE 1999 - Merit certificate from Central Board of Secondary Education, India ______________________________________________________________________________ Other Activities National Service Society (India) ACM Students Chapter, Anna University Computer Science Engineers Association, Anna University ______________________________________________________________________________ Personal Information Country of Citizenship: India US Visa Status: F1 ______________________________________________________________________________ References Available on request.