abhinav sarje

This page gives information about my present academic and research work.

Presently, I am a doctoral candidate of computer engineering at the department of ECpE in Iowa State University. My studies are supported with a research assistantship under my advisor, Dr. Srinivas Aluru. In our research group, we work mostly on parallel and high-performance computing with applications to computational biology and scentific computing.

In the rest of this page, a brief overview of some of my research and course work at Iowa State University is given.

select research projects

A Cloud Computing based abstract framework for computations on Trees On the abstraction lines of MapReduce paradigm, we propose and demonstrate an abstract framework to perform computations on Tree structures. The application specific information is provided by the user and this is neatly abstracted away from the system implementation which may be in a distributed/parallel fashion. More details soon!

Parallel algorithms for sequence alignments on the Cell processor Two prominent parallel algorithms for sequence alignments exist. One uses the wavefront pattern, and the other is based on parallel-prefix. A comparison of these algorithms for performance on a cheap high-performance processor, the Cell BE, leads to a hybrid algorithm based on these two approaches. This scheme gives speedups of four on a Playstation 3 (six SPEs) compared to a Pentium 4 processor. Furthermore, it gives speedups of seven on a QS20 Cell blade (sixteen SPEs). These hold for the three varieties of alignments – global/local, spliced and syntenic.

Construction of gene regulatory networks in parallel on a cluster of Cell processors A parallel algorithm for construction of gene regulatory network, based on mutual information computations using B-splines, implemented on a small cluster of Cell processors gives performance comparable to a high-end massive parallel processing system, like the IBM Blue Gene/L. One QS20 Cell blade gives a speedup of more than seven relative to a Pentium D processor. Using a cluster of eight QS20 Cell blades gives a performance gain of two over 128 nodes of Blue Gene/L.

Scheduling pairwise computations on Cell processors Given the architectural limitations of the Cell processors -- small Local Stores on the SPEs and explicit DMA transfers, this parallel approach to efficiently schedule all-pairs computations between vectors, which may or may not fit in the LS, obtains its efficiency and optimality by minimizing the number of DMA transfers. This approach can be easily applied to dense matrix computations based on pairwise calculations. Applications of this lie in various fields, including materials science, fluid dynamics and systems biology. Our library, libpnorm, is available under LGPL and BSL, and can be downloaded from our group website.

select publications

Abhinav Sarje Jaroslaw Zola Srinivas Aluru Accelerating Pairwise Computations on Cell Processors IEEE Transactions on Parallel and Distributed Systems (TPDS) – Special Issue on HighPerformance Computing with Accelerators to appear 2010 joint first authors

Jaroslaw Zola Maneesha Aluru Abhinav Sarje Srinivas Aluru Parallel Information Theory Based Construction of Genome-wide Gene Regulatory Networks IEEE Transactions on Parallel and Distributed Systems (TPDS) to appear 2010

Abhinav Sarje Srinivas Aluru Parallel Genomic Alignments on the Cell Broadband Engine IEEE Transactions on Parallel and Distributed Systems (TPDS) vol. 20, no. 11, pp. 1600-1610 2009

Abhinav Sarje Jaroslaw Zola Srinivas Aluru Constructing Gene Regulatory Networks on Clusters of Cell Processors Proceedings of the 38th International Conference on Parallel Processing (ICPP'09) pp. 108-115 2009 joint first authors

Abhinav Sarje Srinivas Aluru Parallel Biological Sequence Alignments on the Cell Broadband Engine Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium (IPDPS'08) pp. 1-11 2008

Abhinav Sarje Amit Chawre Shivashankar Nair Reinforcement Learning of Player Agents in RoboCup Soccer Simulation Proceedings of the 4th IEEE International Conference on Hybrid Intelligent Systems (HIS'04) pp. 480-481 2004

coursework

fall '05

CPRE 548 Fundamental Algorithms in Computational Biology Dr. Srinivas Aluru
CPRE 594 Probabilistic Methods in Computer Engineering Dr. Srikanta Tirthapura

spring '06

COMS 518 Introduction to Computational Geometry Dr. Yan-Bin Jia
COMS 531 Theory of Computation Dr. Pavan Aduri

fall '06

CPRE 526 Parallel Algorithms and Programming Dr. Srikanta Tirthapura
COMS 562 Database Implementation Dr. Shashi Gadia

spring '07

CPRE 626 Parallel Algorithms for Scientific Applications Dr. Srinivas Aluru
CPRE 540 Principles of Compiling Dr. Hui-Hsien Chou

fall '07

CPRE 581 Computer Systems Architecture Dr. Zhao Zhang

spring '08

COMS 612 Distributed Algorithms Dr. Soma Chaudhuri

aboutcopyright © Abhinav Sarje 2010