liu.seSearch for publications in DiVA
Change search
Link to record
Permanent link

Direct link
BETA
Östlund, Per
Publications (2 of 2) Show all publications
Pop, A., Östlund, P., Casella, F., Sjölund, M. & Franke, R. (2019). A New OpenModelica Compiler High Performance Frontend. In: Anton Haumer: OTH Regensburg, Germany (Ed.), Proceedings of the 13th International Modelica Conference, Regensburg, Germany, March 4–6, 2019: . Paper presented at The 13th International Modelica Conference, Regensburg, Germany, March 4–6, 2019. Linköping
Open this publication in new window or tab >>A New OpenModelica Compiler High Performance Frontend
Show others...
2019 (English)In: Proceedings of the 13th International Modelica Conference, Regensburg, Germany, March 4–6, 2019 / [ed] Anton Haumer: OTH Regensburg, Germany, Linköping, 2019Conference paper, Published paper (Refereed)
Abstract [en]

The equation-based object-oriented Modelica language allows easy composition of models from components. It is very easy to create very large parametrized models using component arrays of models. Current open-source and commercial Modelica tools can with ease handle models with a hundred thousand equations and a thousand states. However, when the system size goes above half a million (or more) equations the tools begin to have problems with scalability. This paper presents the new frontend of the OpenModelica compiler, designed with scalability in mind. The new OpenModelica frontend can handle much larger systems than the current one with better time and memory performance. The new frontend was validated against large models from the ScalableTestSuite library and Modelica Standard Library, with good results.

Place, publisher, year, edition, pages
Linköping: , 2019
Series
Linköping Electronic Conference Proceedings, ISSN 1650-3686, E-ISSN 1650-3740 ; 157
Keywords
OpenModelica, compiler, flattening, frontend, modelling, simulation, equation-based, scalability
National Category
Computer Systems
Identifiers
urn:nbn:se:liu:diva-154345 (URN)10.3384/ecp19157689 (DOI)978-91-7685-122-7 (ISBN)
Conference
The 13th International Modelica Conference, Regensburg, Germany, March 4–6, 2019
Funder
VINNOVA, 2015-03027
Available from: 2019-02-06 Created: 2019-02-06 Last updated: 2019-02-14Bibliographically approved
Östlund, P., Stavåker, K. & Fritzson, P. (2010). Parallel Simulation of Equation-Based Models on CUDA-Enabled GPUs. In: Proceedings of the 9th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing POOSC '10: . Paper presented at 9th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing (POOSC'10), October 18, 2010, Renoe-Tahoe Nevada, USA. New York, USA: ACM
Open this publication in new window or tab >>Parallel Simulation of Equation-Based Models on CUDA-Enabled GPUs
2010 (English)In: Proceedings of the 9th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing POOSC '10, New York, USA: ACM , 2010Conference paper, Published paper (Refereed)
Abstract [en]

Our contributions with this work are methods and a prototype implementation for compiling and executing a limited set of equation-based mathematical models (written in the object-oriented equation-based modeling language Modelica) on CUDA-enabled GPUs. We look at methods of finding parallelism in Modelica models, that can be used on the massively parallel CUDA architecture. The methods have been implemented in a new back-end module of the OpenModelica compiler (an open-source Modelica compiler). This paper shows that it is possible to automatically generate simulation code for pure continuous-time models that can be reduced to an ordinary differential equation system without algebraic loops and where the initial values of all variables and parameters are known at compile time. It is possible to get some speedup compared with simulation on a single CPU core, a (approximated) relative speedup of 4.6 was for instance obtained for one model.

Place, publisher, year, edition, pages
New York, USA: ACM, 2010
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:liu:diva-110243 (URN)10.1145/2039312.2039317 (DOI)978-1-4503-0546-4 (ISBN)
Conference
9th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing (POOSC'10), October 18, 2010, Renoe-Tahoe Nevada, USA
Available from: 2014-09-04 Created: 2014-09-04 Last updated: 2015-05-18Bibliographically approved
Organisations

Search in DiVA

Show all publications