A Novel Coupling Pattern in Computational Science and Engineering Software

Computational science and engineering (CSE) software is written by experts of certain area(s). Due to the specialization, existing CSE software may need to integrate other CSE software systems developed by different groups of experts. The coupling problem is one of the challenges for software integr...

Full description

Saved in:
Bibliographic Details
Published in:2017 IEEE/ACM 12th International Workshop on Software Engineering for Science (SE4Science) pp. 9 - 12
Main Authors: Kamble, Suraj, Xiaoyu Jin, Nan Niu, Simon, Michelle
Format: Conference Proceeding
Language:English
Published: IEEE 01.05.2017
Subjects:
Calibration
Computational modeling
Computational science and engineering software
coupling pattern
Couplings
Parameter estimation
Semantics
software integration
Software systems
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Computational science and engineering (CSE) software is written by experts of certain area(s). Due to the specialization, existing CSE software may need to integrate other CSE software systems developed by different groups of experts. The coupling problem is one of the challenges for software integration. Here, the coupling we study means the issues involved in integrating 2 legacy codes together, but not the multiphysics coupling where distinct codes are combined in order to model particular phenomena. In this paper, we identify a complex coupling pattern when trying to integrate two CSE software systems. We describe the coupling pattern in detail and show the complexity of resolving such kind of coupling patterns. Our work contributes to area of CSE software since there were few previous studies addressing the coupling problem in CSE domain. Our work will further inspire future research in solving the coupling problem during CSE software integration.
DOI:10.1109/SE4Science.2017.10