About Me
I joined Cisco Systems in June of 2009 in Knoxville, TN.
A copy of my final thesis can be found here.
I received my M.S. in Computer Science from the Virginia Polytechnic Institute and State University in May 2009. I received my B.S. in Software Engineering from the Pennsylvania State University in May of 2007.
While at Virginia Tech, my research was under the guidance of Dr. Dimitris Nikolopoulos and my interests include:
- Operating System Schedulers for Asymmetic Multi-core Systems
- Emerging Parallel Architectures: CMP/SMP Multiprocessors, Cell BE
- Data distribution in a cluster of small-memory nodes
Publications
- Scott Schneider, Jae-Seung Yeom, Benjamin Rose, John C. Linford, Adrian Sandu, Dimitrios S. Nikolopoulos. A Comparison of Programming Models for Multiprocessors with Explicitly Managed Memory Hierarchies. Proc. of the 14th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Raleigh, NC, February 2009.
- Mustafa Rafique, Benjamin Rose, Ali R. Butt, Dimitrios S. Nikolopoulos. Supporting MapReduce on Asymmetric Multicore Clusters. ACM Operating Systems Review, to appear, April 2009.
- Mustafa Rafique, Benjamin Rose, Ali R. Butt, Dimitrios S. Nikolopoulos. CellMR: A Framework for Supporting MapReduce on Asymmetric Cell-Based Clusters. Proc. of the 23rd IEEE International Parallel and Distributed Processing Symposium (IPDPS), Rome, Italy, May 2009.
- Filip Blagojevic, Costin Iancu, Katherine A. Yelick, Dimitrios S. Nikolopoulos, Benjamin Rose and Matthew Curtis-Maury. Scheduling Dynamic Parallelism on Accelerators. 2009 ACM Conference on Computing Frontiers, Ischia, Italy, May 2009.
CV/Résumé
A web-only version of my Résumé can be found here. Please e-mail me for further contact information.
Projects
- CellStream - A framework written for the Cell BE that can stream data between the SPEs and/or main memory at very fast speeds. Designed to allow a user to drop in a computational kernel in all/any SPEs.
- CellMR - A framework implementing MapReduce on multiple accelerator nodes driven by a single traditional multi-core head node.(For example: An Opteron system connected to multiply Playstation 3s via Ethernet)
- Exploration of multiple layers of data parallelism for clusters of asymmetric multi-core processors - Poster of current research work for CS@VT symposium.