skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Nebo: An efficient, parallel, and portable domain-specific language for numerically solving partial differential equations

Authors:
; ; ; ORCiD logo
Publication Date:
Sponsoring Org.:
USDOE
OSTI Identifier:
1398661
Grant/Contract Number:
AC52-07NA27344
Resource Type:
Journal Article: Publisher's Accepted Manuscript
Journal Name:
Journal of Systems and Software
Additional Journal Information:
Journal Volume: 125; Journal Issue: C; Related Information: CHORUS Timestamp: 2017-10-08 15:06:33; Journal ID: ISSN 0164-1212
Publisher:
Elsevier
Country of Publication:
United States
Language:
English

Citation Formats

Earl, Christopher, Might, Matthew, Bagusetty, Abhishek, and Sutherland, James C. Nebo: An efficient, parallel, and portable domain-specific language for numerically solving partial differential equations. United States: N. p., 2017. Web. doi:10.1016/j.jss.2016.01.023.
Earl, Christopher, Might, Matthew, Bagusetty, Abhishek, & Sutherland, James C. Nebo: An efficient, parallel, and portable domain-specific language for numerically solving partial differential equations. United States. doi:10.1016/j.jss.2016.01.023.
Earl, Christopher, Might, Matthew, Bagusetty, Abhishek, and Sutherland, James C. Wed . "Nebo: An efficient, parallel, and portable domain-specific language for numerically solving partial differential equations". United States. doi:10.1016/j.jss.2016.01.023.
@article{osti_1398661,
title = {Nebo: An efficient, parallel, and portable domain-specific language for numerically solving partial differential equations},
author = {Earl, Christopher and Might, Matthew and Bagusetty, Abhishek and Sutherland, James C.},
abstractNote = {},
doi = {10.1016/j.jss.2016.01.023},
journal = {Journal of Systems and Software},
number = C,
volume = 125,
place = {United States},
year = {Wed Mar 01 00:00:00 EST 2017},
month = {Wed Mar 01 00:00:00 EST 2017}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record at 10.1016/j.jss.2016.01.023

Save / Share:
  • A multilevel wavelet collection method for the solution of partial differential equations is developed. Two different approaches of treating general boundary conditions are suggested. Both are based on the wavelet interpolation technique developed in the present research. The first approach uses wavelets as a basis and results in a differential-algebraic system of equations, where the algebraic part arises from boundary conditions. The second approach utilizes extended wavelets, which satisfy boundary conditions exactly. This approach results in a system of coupled ordinary differential equations. The method is tested on the one-dimensional Burgers equation with small viscosity. The solutions are compared withmore » those resulting from the use of other numerical algorithms. The present results indicate that the method is competitive with well-established numerical algorithms. 24 refs., 8 figs., 5 tabs.« less
  • A dynamically adaptive multilevel structure of the algorithm provides a simple way to adapt computational refinements to local demands of the solution. High resolution computations are performed only in regions where sharp transitions occur. The scheme handles general boundary conditions. The method is applied to the solution of the one-dimensional Burgers equation with small viscosity, a moving shock problem, and a nonlinear thermoacoustic wave problem. The results indicate that the method is very accurate and efficient. 16 refs., 9 figs., 2 tab.
  • The standard numerical algorithms for solving parabolic partial differential equations are inherently sequential in the time direction. This paper describes an algorithm for the time-accurate solution of certain classes of parabolic partial differential equations that can be parallelized in both time and space. It has a serial complexity that is proportional to the serial complexities of the best-known algorithms. The algorithm is a variant of the multigrid waveform relaxation method where the scalar ordinary differential equations that make up the kernel of computation are solved using cyclic reduction-type algorithm. Experimental results obtained on a massively parallel multiprocessor are presented.
  • The authors discuss the parallel implementation of preconditioned conjugate gradient (PCG)-based domain decomposition techniques for self-adjoint elliptic partial differential equations in two dimensions on several architectures. The complexity of these methods is described on a variety of message-passing parallel computers as a function of the size of the problem, number of processors and relative communication speeds of the processors. They show that communication startups are very important, and that even the small amount of global communication in these methods can significantly reduce the performance of many message-passing architectures.
  • In this paper, the authors review several methods for solving large sparse linear systems arising from discretization of elliptic partial differential equations on parallel computers. The main idea is to use domain decomposition to introduce some parallelism. The authors describe techniques such as the Schwartz algorithm and the block incomplete factorization method, showing with examples the efficiency of these domain decomposition techniques when the number of subdomains is increased.