Workshop Schedule

Sunday, December 5th, 2010

8:30 – 10:10am Session 1

Opportunities for Leveraging OS Virtualization in High-End Supercomputing
Kevin Pedretti, Sandia National Lab
Patrick G. Bridges, University of New Mexico

Macroarchitecture: A Unifying Framework for Manycore Architecture Research
Ben Cordes, Northeastern University
Gunar Schirner, Northeastern University
David Kaeli, Northeastern University

Development of Virtual CUDA Systems for Parallel Processing on CPU and GPU
Doohwan Oh, Yonsei University, Korea
Sangpil Lee, Yonsei University, Korea
Deokho Kim, Yonsei University, Korea
Changmin Lee, Yonsei University, Korea
Won W. Ro, Yonsei University, Korea

Can High-Performance Interconnects Benefit Hadoop Distributed File System?
Sayantan Sur, Ohio State University
Hao Wang, Ohio State University
Jian Huang, Ohio State University
Xiangyong Ouyang, Ohio State University
Dhabaleswar Panda, Ohio State University  

10:10 – 10:30am Coffee break

10:30 – 12:10pm Session 2

Performance and Scaling Impacts of Hardware-accelerated I/O Virtualization
Hussam Mousa, Intel Corporation
Kshitij A. Doshi, Intel Corporation
Timothy Sherwood, UCSB

PolyHype: Towards Poly-Hypervisor Platform for Cloud Computing
Taeweon Suh, Korea University, Korea
Jong Beom Lim, Korea University, Korea
Dong Hyuk Woo, Intel Labs
Weidong Shi, Nokia Research Center
Xinwen Zhang, Huawei USA
Jong Hyuk Lee, Korea University, Korea

Cooperative Virtual Machine Scheduling on Multi-core Multi-threading Systems - A Feasibility Study
Dulcardo Arteaga, Florida International University
Ming Zhao, Florida International University
Chen Liu, Florida International University
Pollawat Thanarungroj, Florida International University
Lichen Weng, Florida International University

An Optimized I/O Requests Scheme for Network Virtualization
Dongxiao Xu, Intel China Software Center
Kai Chen, Shanghai Jiao Tong University
Eddie Dong, Intel China Software Center
Zhichen Ni, Shanghai Jiao Tong University
Alei Liang, Shanghai Jiao Tong University
Haibing Guan, Shanghai Jiao Tong University

* 25min for each presentation