High Performance Computing

In the area of high performance computing, problems of interest are to arrive at parallel algorithms and implementation on current architectures such as the GPUs, IBM Cell, multicore architectures, etc.

The list of problems that are being investigated presently are: 1. Graph algorithms for GPUs 2. Analytical modeling of GPUs 3. Hybrid algorithms

Major Funded Projects

1. IBM Hybrid Multicore Center of Excellence

Achievements

Others

1. Presented a tutorial titled ''GPU Architectural, Programming, and Performance Models'', at ACM Symposium on Principles and Practice of Parallel Programming (PPoPP), January 2010, Bangalore, India.
2. Tutorial titled ''The GPGPU Phenomenon : Understanding its Scope, Applicability, and its Limitations'', to be presented at ICDCN, 2011, Bangalore, India, January 2011.

Conferences Organized

1. A Workshop on Multicore Technologies was organized during July 23--25, 2009.
2. Workshop on Hybrid Multicore Computing, December 19, 2010. To be held in conjunction with the IEEE International High Performance Computing Conference (HiPC), 2010.

Faculty

1. Kishore Kothapalli (Head)

Students

1. Dip Shankar Banerjee
2. Kiran Kumar Matham