Stony Brook University Logo Computer Science

Rezaul Chowdhury

Back to Faculty Directory

Rezaul Chowdhury
Assistant Professor
Computer Science Department
1421 Computer Science
Stony Brook University
Stony Brook, NY 11794-4400

Phone: 631-632-8959
Website: http://www.cs.stonybrook.edu/~rezaul
Email

INTERESTS

Algorithms and Data Structures for Efficient Serial and Parallel Computations, Cache- and I/O-efficient Computing, Computational Biology and Bioinformatics, Experimental Algorithmics.

BIOGRAPHY

Rezaul Chowdhury received his Ph.D. from the Department of Computer Sciences, UT Austin, working with Professor Vijaya Ramachandran, and defending "Cache-efficient Algorithms and Data Structures: Theory and Experimental Evaluation". Prior to joining SBU in 2011, he was in Boston working with Professor Sandor Vajda's Structural Bioinformatics Group at Boston University, and Professor Charles Leiserson's SuperTech Research Group at MIT. Before moving to Boston, he was a postdoctoral fellow at the Center for Computational Visualization (CVC), Institute for Computational Engineering & Sciences (ICES), University of Texas at Austin. He worked with Professor Chandrajit Bajaj. Chowdhury now leads the Theoretical and Experimental Algorithmics (TEA) Group where they concentrate on both algorithm design and algorithm engineering. He holda a joint appointment with the Institute for Advanced Computational Sciences (IACS).

RESEARCH

Rezaul Chowdhury is currently involved in an NSF-funded project with Charles Leiserson and Steven Johnson of MIT on building a stencil computation compiler called "Pochoir". A stencil defines the value of a grid point in a d-dimensional spatial grid at time t as a function of neighboring grid points at recent times before t. Stencil computations are conceptually simple to implement using nested loops, but looping implementations suffer from poor cache performance on multicore processors. Cache-efficient divide-and-conquer stencil algorithms exist, but most programmers find them difficult to implement. Moreover, open problems remain in adapting these algorithms to realistic applications that lack the perfect regularity of simple examples. This research aims to develop a language embedded in C++ that can express stencil computations concisely and can be compiled automatically into highly efficient algorithmic code that will make good use of the memory hierarchy and processor pipelines endemic to multicore processors and will run fast on a diverse set of hardware platforms. A wide variety of stencil-based applications — ranging across physics, biology, chemistry, energy, climate, mechanical and electrical engineering, finance, and other areas — will become easier to develop and maintain.

Some of his other projects focus on efficient resource-oblivious algorithms, fast energetics computation, and protein-protein docking.

AWARDS & ACTIVITIES

Rezaul Chowdhury received a best paper award in IPDPS 2010 for introducing the notion of multicore-oblivious algorithms. He is also interested in programming contests, and won an ACM ICPC Regional Contest as a student. Some contest problems he authored for training students are included in the "Programming Challenges: The Programming Contest Training Manual" by Steven Skiena & Miguel Revilla.

TEACHING SUMMARY

CSE 548, CSE 590, CSE 613, CSE 638, CSE 642