RT Dissertation/Thesis
T1 High Performance Computing for Genomics
T2 Computación de alto rendimiento para la genómica
A1 Perez-Wohlfeil, Esteban
K1 Genómica
K1 Secuencia nucleotídica
K1 Arquitectura de ordenadores - Tesis doctorales
AB With the thrive of data acquisition methods, computerized research has become increasingly more common. However, in order to match the huge data-processing demands, the design of new algorithms along with their optimization on specific hardware platforms has become a necessity. This scenario is particularly true in the case of comparative genomics, where massive DNA sequences are being published daily, and their processing presents many computational bottlenecks.The comparison of DNA sequences is a central problem with direct impact on human health, and therefore its computational acceleration is of wide interest. However, due to its arbitrary nature, the parallel acceleration of sequence comparison poses computational challenges such as including heterogeneous granularity, unpredictable load, etc. In order to achieve high performance, algorithms must be tailored to the underlying hardware model, which may represent different computational approaches and often require even the redesign of the algorithms themselves.This thesis addresses a computational tour of the sequence comparison problem by making use of hardware and algorithmic optimizations in single core machines, shared memory systems and Graphic Processing Units. The first contribution features a formal framework that enables unlimited search space size in strictly linear time. The second contribution describes a parallelization using shared memory machines that achieves high sensitivity in metagenomic sequences. The third contribution describes the overcoming of the data parallelism model in GPUs for the irregular pairwise sequence comparison. Lastly, the use of Machine-Learning-aided schedulers is explored to improve resource allocation and throughput in supercomputers dedicated to sequence comparison.
PB UMA Editorial
YR 2023
FD 2023
LK https://hdl.handle.net/10630/27835
UL https://hdl.handle.net/10630/27835
LA eng
NO The results are twofold: from a computational perspective, new High Performance Computing methodologies and parallel mechanisms are proposed for different hardware architectures; and from a comparative genomics perspective, the complexity of sequence comparison has been lowered while providing exhaustive and heuristic approaches that can be run on both commodity and specialized hardware.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026