Testing Weighted Splitting Schemes on a One-Column Transport-Chemistry Model

Mike Botchev, I. Faragó, Ágnes Havasi

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In many transport-chemistry models, a huge system of ODE's of the advection-diffusion-reaction type has to be integrated in time. Typically, this is done with the help of operator splitting. Rosenbrock schemes combined with approximate matrix factorization (ROS-AMF) are an alternative to operator splitting which does not suffer from splitting errors. However, implementation of ROS-AMF schemes often requires serious changes in the code. In this paper we test another classical second order splitting introduced by Strang in 1963, which, unlike the popular Strang splitting, seemed to be forgotten and rediscovered recently (partially due to its intrinsic parallellism). This splitting, called symmetrically weighted sequential (SWS) splitting, is simple and straightforward to apply, independent of the order of the operators and has an operator-level parallelism. In the experiments, the SWS scheme compares favorably to the Strang splitting, but is less accurate than ROS-AMF.

Original languageEnglish
Pages (from-to)295-302
Number of pages8
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2907
Publication statusPublished - 2004

Fingerprint

Factorization
Chemistry
Approximate Factorization
Testing
Matrix Factorization
Operator Splitting
Strings
Advection
Mathematical operators
Model
Advection-diffusion
Operator
Parallelism
Experiments
Alternatives
Experiment

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

@article{0697d742535e4ba2b0bda9cd7db5f03a,
title = "Testing Weighted Splitting Schemes on a One-Column Transport-Chemistry Model",
abstract = "In many transport-chemistry models, a huge system of ODE's of the advection-diffusion-reaction type has to be integrated in time. Typically, this is done with the help of operator splitting. Rosenbrock schemes combined with approximate matrix factorization (ROS-AMF) are an alternative to operator splitting which does not suffer from splitting errors. However, implementation of ROS-AMF schemes often requires serious changes in the code. In this paper we test another classical second order splitting introduced by Strang in 1963, which, unlike the popular Strang splitting, seemed to be forgotten and rediscovered recently (partially due to its intrinsic parallellism). This splitting, called symmetrically weighted sequential (SWS) splitting, is simple and straightforward to apply, independent of the order of the operators and has an operator-level parallelism. In the experiments, the SWS scheme compares favorably to the Strang splitting, but is less accurate than ROS-AMF.",
author = "Mike Botchev and I. Farag{\'o} and {\'A}gnes Havasi",
year = "2004",
language = "English",
volume = "2907",
pages = "295--302",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Testing Weighted Splitting Schemes on a One-Column Transport-Chemistry Model

AU - Botchev, Mike

AU - Faragó, I.

AU - Havasi, Ágnes

PY - 2004

Y1 - 2004

N2 - In many transport-chemistry models, a huge system of ODE's of the advection-diffusion-reaction type has to be integrated in time. Typically, this is done with the help of operator splitting. Rosenbrock schemes combined with approximate matrix factorization (ROS-AMF) are an alternative to operator splitting which does not suffer from splitting errors. However, implementation of ROS-AMF schemes often requires serious changes in the code. In this paper we test another classical second order splitting introduced by Strang in 1963, which, unlike the popular Strang splitting, seemed to be forgotten and rediscovered recently (partially due to its intrinsic parallellism). This splitting, called symmetrically weighted sequential (SWS) splitting, is simple and straightforward to apply, independent of the order of the operators and has an operator-level parallelism. In the experiments, the SWS scheme compares favorably to the Strang splitting, but is less accurate than ROS-AMF.

AB - In many transport-chemistry models, a huge system of ODE's of the advection-diffusion-reaction type has to be integrated in time. Typically, this is done with the help of operator splitting. Rosenbrock schemes combined with approximate matrix factorization (ROS-AMF) are an alternative to operator splitting which does not suffer from splitting errors. However, implementation of ROS-AMF schemes often requires serious changes in the code. In this paper we test another classical second order splitting introduced by Strang in 1963, which, unlike the popular Strang splitting, seemed to be forgotten and rediscovered recently (partially due to its intrinsic parallellism). This splitting, called symmetrically weighted sequential (SWS) splitting, is simple and straightforward to apply, independent of the order of the operators and has an operator-level parallelism. In the experiments, the SWS scheme compares favorably to the Strang splitting, but is less accurate than ROS-AMF.

UR - http://www.scopus.com/inward/record.url?scp=33745311995&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745311995&partnerID=8YFLogxK

M3 - Article

VL - 2907

SP - 295

EP - 302

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -