GPU accelerated quantum chemistry

Placeholder Show Content

Abstract/Contents

Abstract
This dissertation develops techniques to accelerate quantum chemistry calculations using commodity graphical processing units (GPUs). As both the principle bottleneck in finite basis calculations and a highly parallel task, the evaluation of Gaussian integrals is a prime target for GPU acceleration. Methods to tailor quantum chemistry algorithms from the bottom up to take maximum advantage of massively parallel processors are described. Special attention is taken to make maximum use of performance features typical of modern GPUs, such as high single precision performance. After developing an efficient integral direct self-consistent field (SCF) procedure for GPUs that is an order of magnitude faster than typical CPU codes, the same machinery is extended to the configuration interaction singles (CIS) and time- dependent density functional theory (TDDFT) methods. Finally, this machinery is applied to molecular dynamics (MD) calculations. To extend the time scale accessible to MD calculations of large systems, an ab initio multiple time steps (MTS) approach is developed. For small systems, up to a few dozen atoms, an interactive interface enabling a virtual molecular modeling kit complete with realistic ab initio forces is developed.

Description

Type of resource text
Form electronic; electronic resource; remote
Extent 1 online resource.
Publication date 2015
Issuance monographic
Language English

Creators/Contributors

Associated with Luehr, Nathan
Associated with Stanford University, Department of Chemistry.
Primary advisor Martinez, Todd J. (Todd Joseph), 1968-
Thesis advisor Martinez, Todd J. (Todd Joseph), 1968-
Thesis advisor Andersen, Hans, 1941-
Thesis advisor Pande, Vijay
Advisor Andersen, Hans, 1941-
Advisor Pande, Vijay

Subjects

Genre Theses

Bibliographic information

Statement of responsibility Nathan Luehr.
Note Submitted to the Department of Chemistry.
Thesis Thesis (Ph.D.)--Stanford University, 2015.
Location electronic resource

Access conditions

Copyright
© 2015 by Nathan Luehr
License
This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).

Also listed in

Loading usage metrics...