Cy Chan

I recently completed my PhD in the Electrical Engineering and Computer Science Department and CSAIL at MIT under the supervision of Alan Edelman. I am currently a computational science postdoctoral research fellow at the Lawrence Berkeley National Laboratory. My research interests are in high performance parallel computing and algorithmic optimization of applications for energy efficiency (efficient combustion) and renewable energy. I have contributed benchmarks and algorithmic development to the PetaBricks project in the COMMIT group. I have also done work in random matrix theory and applications.

Publications:

• Jason Ansel, Yee Lok Wong, Cy Chan, Marek Olszewski, Alan Edelman, and Saman Amarasinghe, "Language and Compiler Support for Auto-Tuning Variable-Accuracy Algorithms," Proceedings of the International Symposium on Code Generation and Optimization (CGO), April 2011. [pdf]
• Jason Ansel and Cy Chan, "PetaBricks: Building adaptable and more efficient programs for the multi-core era," Crossroads, The ACM Magazine for Students (XRDS), Volume 17, Issue 1, September 2010. [pdf]
• Shoaib Kamil, Cy Chan, Leonid Oliker, John Shalf, Samuel Williams, "An Auto-Tuning Framework for Parallel Multicore Stencil Computations," Proceedings of the IEEE International Parallel & Distributed Processing Symposium (IPDPS), 2010. [pdf]
• Cy Chan, Jason Ansel, Yee Lok Wong, Saman Amarasinghe, Alan Edelman, "Autotuning Multigrid with PetaBricks," Proceedings of the International Conference for High Performance Computing, Networking, Storage, and Analysis (SC), 2009. [pdf]
• Shoaib Kamil, Cy Chan, Samuel Williams, Leonid Oliker, John Shalf, Mark Howison, E. Wes Bethel, Prabhat, "A Generalized Framework for Auto-tuning Stencil Computations", Cray User Group (CUG), 2009. (WINNER: Best paper) [pdf]
• Jason Ansel, Cy Chan, Yee Lok Wong, Marek Olszewski, Qin Zhao, Alan Edelman, Saman Amarasinghe, "PetaBricks: A Language and Compiler for Algorithmic Choice," Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), 2009. [pdf]
• Cy Chan, Vesselin Drensky, Alan Edelman, Raymond Kan, and Plamen Koev, "On Computing Schur Functions and Series Thereof," To appear in the Journal of Algebraic Combinatorics. [pdf]
• James Albrecht, Cy Chan, and Alan Edelman, "Sturm Sequences and Random Eigenvalue Distributions," Foundations of Computational Mathematics, 2008. [pdf, web]
• Preston A Jackson, Cy Chan, Charles M Rader, M Michael Vai, Jonathan E Scalera, "A Systolic FFT Architecture for Real Time FPGA Systems," Eighth Annual Workshop on High Performance Embedded Computing (HPEC), 2004. [abstract, presentation]
• Gil Raz, Cy Chan, "Method and System of Nonlinear Signal Processing," US Patent 7609759. [pdf, web]

Reports and Papers:

• Cy Chan, Pamela Silva, and Chao Zhang, "Danehy Park Wind Turbine Project -- Preliminary Assessment Report," Project report submitted to the City of Cambridge, 2011. [pdf, web]
• Jason Ansel, Yee Lok Wong, Cy Chan, Marek Olszewski, Alan Edelman, and Saman Amarasinghe, "Language and Compiler Support for Auto-Tuning Variable-Accuracy Algorithms," Technical Report MIT-CSAIL-TR-2010-032, July 2010. [pdf]
• Kathy Araujo, Katherine Dykes, Cy Chan, Alex Kalmikov, Kevin Ferrigno, Bryan Palmintier, Sungho Lee, Mark Lipson, "Feasibility Study -- Project Full Breeze," Project report submitted to MIT Facilities, 2010. [pdf, web]
• Cy Chan, James Stalker, Alan Edelman, Stephen Connors, "Leveraging High Performance Computation for Statistical Wind Prediction," American Wind Energy Association Wind Power Conference and Exhibition (WINDPOWER), 2010. [pdf]
• Alex Kalmikov, Guillaume Dupont, Katherine Dykes, Cy Chan, "Wind Power Resource Assessment in Complex Urban Environments: MIT Campus Case-Study Using CFD Analysis," American Wind Energy Association Wind Power Conference and Exhibition (WINDPOWER), 2010. [pdf]

Posters and Presentations:

• "Cambridge Danehy Park Wind Turbine Preliminary Project Assessment," MIT Energy Night, October 2011. [pdf]
• "Resource Assessment Overview and MIT Full Breeze Case Study," MIT Wind Week, January 2011. [pdf]
• "Leveraging High Performance Computation for Statistical Wind Prediction," American Wind Energy Association Wind Power Conference and Exhibition (WINDPOWER), 2010. [pdf]
• "Wind Power Resource Assessment in Complex Urban Environments: MIT Campus Case-Study Using CFD Analysis", American Wind Energy Association Wind Power Conference and Exhibition (WINDPOWER), 2010. [pdf]
• "Autotuning Multigrid with PetaBricks," SIAM Conference on Parallel Processing for Scientific Computing (PP), 2010. [pdf]
• "Autotuning Multigrid with PetaBricks," International Conference for High Performance Computing, Networking, Storage, and Analysis (SC), 2009.
• "A Dynamic Programming Approach to Autotuning Multigrid," Fourteenth Copper Mountain Conference on Multigrid Methods, 2009.

Other Projects:

• Efficient Hypergeometric Function Evaluation - In collaboration with Plamen Koev, Raymond Kan, Vesselin Drensky, and Alan Edelman, I've written an efficient software package that evaluates hypergeometric functions of one or two matrix arguments in the case of α = 1. Please see our related paper (noted above).


• Image Segmentation with Matlab Star-P - With Joanne Wong, I wrote a program to do image segmentation using the ITK Segmentation Toolkit on a parallel computer using Matlab Star-P as a front end. This was a project for 18.337 - Applied Parallel Computing in Spring 2006.


• Beta Estimator - I wrote a webpage that does parameter estimation (maximum likelihood) for the eigenvalue spacings distributions of beta-Hermite random matrix ensembles. If you have spacings data and would like to estimate which beta-Hermite ensemble has eigenvalues that best match your distribution, give this a try. This was a project for 18.338 - The Eigenvalues of Random Matrices in Fall 2005.


• Pseudo-Eigenvectors of the β-Hermite Matrix Model - For fun, I made some movies showing the "pseudo-eigenvectors" produced by the "shooting method" for different values of λ in (A-λI)x = 0, for a fixed β-Hermite matrix A.