High Performance Computing Qualifying Exam

Spring 2009
Committee:  Calvin J. Ribbens,  Adrian Sandu (Chair), Eunice E. Santos, Layne T. Watson

The High Performance Computing (HPC) Qualifying Exam has both a written and an oral component.

Written: For each of the papers on the reading list write a short (no more than 1 page) summary and critique of the paper in your own words. Your writeup should identify the main points or contributions of the paper, as well as the significance of the contributions. Highlight any particularly strong or weak aspects of a paper. Submit this written material to Dr. Sandu via email (sandu@cs.vt.edu) by February 1, 2009.

Oral: In mid February, you will make an oral presentation to the committee. From the reading list, choose (roughly) 4 papers that, in your view, are connected by an underlying theme or sub-topic in HPC. The reading list comprises of a wide span of topics in HPC and there are a variety of paper combinations that represent a variety of HPC sub-topics. Prepare roughly 45 minutes of material. Do not provide a tutorial or short course about the details or minutiae in the papers. Your material should


The committee will ask questions throughout the oral exam not only regarding the 3-4 papers you chose, but on any of the papers in the list (and potential research directions).

In preparing for this exam, you should do the work yourself and should find and read any other references that seem useful. You may discuss papers in the reading list with other students who are preparing for the exam. However, you are not to work with others in preparing submissions to the exam (e.g., summaries and critiques, talk preparation, etc.). You may ask us questions about the papers. However, we will only answer specific, narrowly focused questions. We cannot spend a lot of time helping you figure out the paper - that's your job!
 

Reading List


1. Agarwal, R. C., S. M. Balle, F. G. Gustavson, M. Joshi, and P. Palkar, A 3-dimensional approach to parallel matrix multiplication, IBM Journal of Research and Development, Volume 39, Number 5, pp. 1-8, Sept. 1995.

2. van Leer, B., Towards the ultimate conservative difference scheme. IV. A new approach to numerical convection, Journal of Computational Physics, Vol. 23, pp. 276-299, 1977.

3. Bischof, C., Automatic differentiation, tangent linear models, and (pseudo)adjoints, High-Performance Computing in the Geosciences, Vol. 462, F.X. Le Dimet editor, pp. 59-80, Kluwer Academic Publishers, Boston, 1995, QA3 N3 v.462.

4. Dennis., J.E., and More, J.J., Quasi-Newton methods, motivation and theory, SIAM Review, Vol. 19, No. 1, January 1977, pp. 46-89.

5. Elmroth, E., Gustavson, F., Jonsson, I., and Kagstrom, B., Recursive blocked algorithms and hybrid data structures for dense matrix library software, SIAM Review, Vol. 46, No. 1, pp. 3-45, 2004.

6. Bader, D.A. and JaJa, J., SIMPLE: A methodology for programming high performance algorithms on clusters of symmetric multiprocessors (SMPs), Journal of Parallel and Distributed Computing, Vol. 58, No. 1, pp. 92-108, 1999.

7. Shiloach, Y. and Vishkin, U., An O(log n) parallel connectivity algorithm, Journal of Algorithms, Vol. 3, No. 1, pp. 57-67, 1982.

8. Lewis, R.M., Torczon, V., and Trosset, M.W., Direct search methods: then and now, Journal of Computational and Applied Mathematics, Vol. 124, pp. 191-207, 2000.

9. Valsalam, V. and Skjellum, A., A framework for high-performance matrix multiplication based on hierarchical abstractions, algorithms and optimized low-level kernels, Concurrency and Computation: Practice and Experience, Vol. 14 , pp 805-839, 2002.

10. Yang, T., and Gerasoulis, A., DSC: scheduling parallel tasks on an unbounded number of processors, IEEE Transactions on Parallel and Distributed Systems, Vol. 5, No. 9, pp. 951-967, 1994.