Thesis defence E.J. Houtgast: genomics

11 November 2019 10:00 - Location: Aula, TU Delft - By: webredactie

On Hardware-Accelerated Maximally-Efficient Systolic Arrays. Acceleration and Optimization of Genomics Pipelines Through Hardware/Software Co-Design. Promotor 1: Dr.ir. Z. Al-Ars (EWI); Promotor 2: Prof.dr.ir. K.L.M. Bertels (EWI).

Developments in sequencing technology have drastically reduced the cost of DNA sequencing, resulting in its widespread adoption and in an exponential increase in data being generated. Therefore, acceleration and optimization of genomics processing pipelines is becoming increasingly important.

The BWA-MEM genomic mapping algorithm is a critical first step of many genomics pipelines, mapping raw input sequences onto a reference genome. A major part of overall BWA-MEM execution time is spent performing Seed Extension, an algorithm related to the Smith-Waterman pairwise sequence alignment algorithm. The standard approach for the heterogeneous acceleration of the Smith-Waterman algorithm is to map it onto a systolic array architecture.

In this dissertation, techniques to improve the efficiency of systolic arrays are proposed: eliminating the dependence of systolic array efficiency on read sequence length and reference sequence length results in a maximally-efficient systolic array. Implementations on FPGA and GPU result in a threefold kernel-level improvement to execution time and in the first FPGA-accelerated and the first GPU-accelerated implementation of BWA-MEM with an overall end-to-end twofold application-level speedup. Moreover, an FPGA Smith-Waterman implementation using the above efficiency improvements results in a performance of 214 GCUPS with 99.8% efficiency, which is the highest reported efficiency and performance of any FPGA-accelerated Smith-Waterman implementation to date. Finally, various aspects of these designs are evaluated, including power-efficiency and design-time.

More information?

For access to theses by the PhD students you can have a look in TU Delft Repository, the digital storage of publications of TU Delft. Theses will be available within a few weeks after the actual thesis defence.