Files in this item



application/pdfECE499-Sp2014-chen.pdf (1MB)Restricted to U of Illinois
(no description provided)PDF


Title:Hardware Acceleration of Short Read Alignment
Author(s):Chen, Daniel
Contributor(s):Chen, Deming
Short Read Alignment
Hardware Acceleration
Abstract:Short read aligners are tools used to map a person's genome onto the human reference genome. A typical aligner may take around seven hours to map the DNA associated with one single chromosome. I propose porting an aligner to FPGA, which would offer speedup in addition to the familiar features present in the original aligner. Traditionally, aligners have relied on multi-core CPUs for alignment. However, this approach does not fully exploit the absurdly parallel nature of short read alignment. Read alignments have essentially no dependencies, which makes them suitable for parallel processing. There exist several GPU and FPGA aligners, but they do not include the full feature set of their CPU counterparts. Open Computing Language (OpenCL) is utilized for application development, which will allow for the ease of use on both GPU and FPGA. Eventually, the aligner will support the same tasks as the most feature-rich aligners of today. Speedups of 5x up to 50x are expected compared to traditional CPU aligners.
Issue Date:2014-05
Date Available in IDEALS:2014-09-22

This item appears in the following Collection(s)

Item Statistics