A Coarse-Grained Reconfigurable Processor for Sequencing and Phylogenetic Algorithms in Bioinformatics
2011 (English)In: Proceedings: 2011 International Conference on Reconfigurable Computing and FPGAs, ReConFig 2011, 2011, 190-197 p.Conference paper (Refereed)
A coarse-grained reconfigurable processor tailoredfor accelerating multiple bioinformatics algorithms isproposed. In this paper, a programmable and scalablearchitectural platform instantiates an array of coarse grainedlight weight processing elements, which allows arbitrarypartitioning, scheduling schemes and capable of solvingcomplete four popular bioinformatics algorithms: theNeedleman-Wunsch, Smith-Waterman, and HMMER onsequencing, and Maximum Likelihood on phylogenetic. Thekey difference of the proposed CGRA based solution comparedto FPGA and GPU based solutions is a much better match onarchitecture and algorithms for the core computational needs,as well as the system level architectural needs. For the samedegree of parallelism, we provide a 5X to 14X speed-upimprovements compared to FPGA solutions and 15X to 78Xcompared to GPU acceleration on 3 sequencing algorithms. Wealso provide 2.8X speed-up compared to FPGA with the sameamount of core logic and 70X compared to GPU with the samesilicon area for Maximum Likelihood.
Place, publisher, year, edition, pages
2011. 190-197 p.
Bioinformatics, Coarse Grained Reconfigurable Architecture, Needleman Wunsch, Smith Waterman, HMMER, Maximum Likelihood, Phylogenetic Inference, VLSI
Electrical Engineering, Electronic Engineering, Information Engineering
IdentifiersURN: urn:nbn:se:kth:diva-73032DOI: 10.1109/ReConFig.2011.1ScopusID: 2-s2.0-84856915938ISBN: 978-1-4577-1734-5OAI: oai:DiVA.org:kth-73032DiVA: diva2:488500
2011 International Conference on Reconfigurable Computing and FPGAs, ReConFig 2011. Cancun, Quintana Roo. 30 November 2011 - 2 December 2011
QC 201203122012-02-012012-02-012012-03-12Bibliographically approved