Arun Raghavan

PhD Student
Architecture and Compilers Group,
Computer and Information Science,
University of Pennsylvania

email	arraghav at cis dot followed by the university domain name
office	614, Levine Hall, 3330 Walnut Street,
	Philadelphia, PA 19104

I am a PhD student in the Computer and Information Sciences department at the University of Pennsylvania. I am advised by Prof. Milo Martin.

My research interests broadly involve shared-memory multicores and programming models for such architectures. My dissertation research on computational sprinting explores building systems around responsiveness, i.e., making a large amount of computation available to applications for short durations of time. The goal of my research is to greatly improve exisiting apps and open the doors for new apps on mobile devices such as smartphones.
In the past, I have worked on cache coherence protocols and transactional memory.
In the summer of 2008, I interned with the Larrabee architecture group at Intel in Hillsboro, OR.

Collaborators

Rajeev Alur, Colin Blundell, Jyotirmoy Deshmukh, Sela Mador-Haim, Milo Martin, Marios Papaefthymiou, Kevin Pipe, Abhishek Udupa, Thomas Wenisch,

Publications

Hardware Software Testbed for Computational Sprinting
Arun Raghavan, Laurel Emurian, Lei Shao, Marios Papaefthymiou, Kevin P. Pipe, Thomas F. Wenisch and Milo M. K. Martin
Proceedings of the 18th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 2013. (to appear)

Computational Sprinting (best paper award)
Arun Raghavan, Yixin Luo, Anuj Chandawalla, Marios Papaefthymiou, Kevin P. Pipe, Thomas F. Wenisch and Milo M. K. Martin
Proceedings of the 18th International Symposium on High Performance Computer Architecture (HPCA), Feb 2012.
Paper: pdf
Talk: pptx

Token Tenure and PATCH: A Predictive/Adaptive Token Counting Hybrid (journal version)
Arun Raghavan, Colin Blundell, and Milo M. K. Martin
ACM Transactions on Architecture and Compiler Optimization (TACO), September 2010.
Paper: pdf

RetCon: Transactional Repair without Replay
Colin Blundell, Arun Raghavan, and Milo M. K. Martin
Proceedings of the 37th International Symposium on Computer Architecture (ISCA), June 2010.
Paper: pdf

Token Tenure: PATCHing Token Counting Using Directory-Based Cache Coherence
Arun Raghavan, Colin Blundell, and Milo M. K. Martin
Proceedings of the International Symposium on Microarchitecture (MICRO), November 2008.
Paper: pdf
Talk: pptx

Teaching

I was the teaching assistant for CIS 371 in Spring 2008 and CIS 501 in Fall 2007.
I gave a guest lecture on branch prediction for CIS 501 in Fall 2010.

Course work & WPE

CIS 500, Software Foundations (Fall 2006)
CIS 501, Computer Architecture (Fall 2006)
CIS 540, Modern Programming Language Implementation (Fall 2006)
CIS 502, Analysis of Algorithms (Spring 2007)
CIS 505, Software Systems (Spring 2007)
CIS 553, Networked Systems (Fall 2008)
CIS 511, Theory of Computation (Spring 2008)
CIS 665, GPU Programming and Architecture (Spring 2008)
CIS 640, Introduction to Multiprocessor Programming (Spring 2009)
CIS 534, Multicore Programming and Architecture (Spring 2010)
CIS 700, Technology and Policy (Spring 2010)
My WPE-II exam (Penn's version of a depth qualifier) was on Hardware Mechanisms for Deterministic Record and Replay