The purpose of this seminar is to study and explore interesting parallel & distributed algorithms and ideas. We will cover a wide range of topics, such as parallel data structures, models of parallel computation, parallel complexity theory, scalable numerical methods, and randomization techniques for parallel algorithms. While most papers will be from the last five years, a few older papers will be included early in the semester to set the stage for people not familiar with parallel algorithms. It is recommended that such students also register for CS-6230 : High Performance Computing & Parallelization.
We will meet and discuss (typically) one paper a week. If you are registering for credit, you will be expected to present (at least) one paper from the list below. your choice of paper(s) at the earliest. If you intend to register for more than 1 credit, please check with me first.
MEB 3147 (LCR)
Here is a tentative list of papers for discussion (in no particular order). We will likely add additional ones as our discussions progress. It is also unlikely that we will be able to cover all of these and will try and prioritize the ordering.
** Use the discussion board on this page to select papers. **
19 Apr - Jimmy Moore Richard Vuduc, Aparna Chandramowlishwaran, Jee Choi, Murat Guney, and Aashay Shringarpure. 2010. On the limits of GPU acceleration. In Proceedings of the 2nd USENIX conference on Hot topics in parallelism (HotPar’10). USENIX Association, Berkeley, CA, USA, 13-13.