Bachelorarbeit BCLR-2021-63

Bibliograph.
Daten
Kharitenko, Pavel: Coupling Julia-based simulations via preCICE.
Universität Stuttgart, Fakultät Informatik, Elektrotechnik und Informationstechnik, Bachelorarbeit Nr. 63 (2021).
73 Seiten, englisch.
Kurzfassung

The coupling library preCICE allows to couple single-physics solvers to partitioned multi-physics simulations in a black-box fashion. preCICE is a C++ library, but it offers language bindings to access the preCICE API from solvers written in other languages, such as C, Python, Fortran and MATLAB. The Julia Programming Language designed for numerical computing is a strong candidate to be supported by preCICE. While Julia provides a wide set of tools for interfacing with other languages, including C++, porting a library such as preCICE that is made for High Performance Computing and runs on a huge number of processes, requires little to no compromises. Multiple ways of wrapping a C/C++ library are presented and implemented. In addition come Julia’s own features, for example the Distributed base library, that deviate from classic standards of known scientific languages. To test the bindings, two dummy solvers are coupled and presented in an example setup, with an outlook on further development.

Volltext und
andere Links
Volltext
Abteilung(en)Universität Stuttgart, Institut für Parallele und Verteilte Systeme, Simulation großer Systeme
BetreuerUekermann, Jun.-Prof. Benjamin; Desai, Ishaan
Eingabedatum22. Dezember 2021
   Publ. Informatik