The problem is then how to utilize this enormous processing power in a user friendly way. Partitioning needs to be made automatically. Modelica gives good possibility to automatically partition the model equation execution into separate threads since it is a declarative language based on equations.
This paper describes a method to automatically parallelize model equations implemented in Dymola. A speed-up of 3.4 times has been achieved using 4 cores/8 threads.
Keywords: Modelica; Multi-core; Automatic partitioning
Proceedings of the 10th International Modelica Conference; March 10-12; 2014; Lund; Sweden
Aronsson, P., and P. Fritzson (2002). Multiprocessor Scheduling of Simulation Code from Modelica Models. In Proceedings of the 2nd International Modelica Conference, Oberpfaffenhofen, Germany, Mar. 18–19, 2002.
Aronsson P. (2006): Automatic Parallelization of Equation-Based Simulation Programs. Institutionen för datavetenskap, 2006.
Casella, F. (2013): A Strategy for Parallel Simulation of Declarative Object-Oriented Models of Generalized Physical Networks. 5th International Workshop of Equation-Based Object-Oriented Modeling Languages and Tools, April 2013, University of Nottingham, UK. Linköping University Electronic Press. http://www.ep.liu.se/ecp_home/index.en.aspx?issue=084.
Elmqvist H., Mattsson S.E., Olsson H. (2002): New Methods for Hardware-in-the-Loop Simulation of Stiff Models, 2nd International Modelica Conference, DLR, Oberpfaffenhofen, Germany, March 18-19.