Modern programming models for GPU-accelerated heterogeneous supercomputers : computational fluid dynamics and in situ data compression
Abstract/Contents
- Abstract
- This work presented in this thesis is concerned with large-scale computational fluid dynamics simulations and can be divided into two main parts. The first is an investigation into the suitability of asynchronous, task-based parallelism for computational fluid dynamics simulations targeting exascale, heterogeneous architectures. The second part is concerned with the incorporation of data compression algorithms into computational fluid dynamics simulations using interpolative decomposition methods. The first chapter is a brief introduction to computational fluid dynamics, as well as an overview of traditional parallel computing practices in the field. The second chapter details the Legion programming system, which is used in this work to develop asynchronous, task-parallel solvers. Chapter Three is a case study on the implementation of a high-order, Navier-Stokes solver using the Regent programming language and the Legion runtime; this is compared to an existing, data-parallel MPI+Fortran version of the solver. The fourth chapter provides an in-depth discussion of interpolative decomposition methods for data compression, and the fifth chapter presents results for the incorporation of a novel interpolative decomposition algorithm into the Legion flow solver described in Chapter Three. Finally, Chapter Six concludes with "lessons learned" and possible avenues for future work.
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2021; ©2021 |
Publication date | 2021; 2021 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Pacella, Heather Elizabeth |
---|---|
Degree supervisor | Iaccarino, Gianluca |
Thesis advisor | Iaccarino, Gianluca |
Thesis advisor | Aiken, Alexander |
Thesis advisor | Darve, Eric |
Degree committee member | Aiken, Alexander |
Degree committee member | Darve, Eric |
Associated with | Stanford University, Department of Mechanical Engineering |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Heather Pacella. |
---|---|
Note | Submitted to the Department of Mechanical Engineering. |
Thesis | Thesis Ph.D. Stanford University 2021. |
Location | https://purl.stanford.edu/yc068wp9748 |
Access conditions
- Copyright
- © 2021 by Heather Elizabeth Pacella
- 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...