Acceleration of FMU Co-Simulation On Multi-core Architectures - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2016

Acceleration of FMU Co-Simulation On Multi-core Architectures


The design of cyber-physical systems is a complex process and relies on the simulation of the system behavior before its deployment. Co-simulation allows system designers to simulate a whole system composed of a number of interconnected subsystems. Traditionally, these models are modeled by experts of different fields using different tools, and then integrated into one environment to perform simulation at the system-level. This results in complex and heavy co-simulations and requires adequate solutions and tools in order to reduce the execution time. Unfortunately, most modeling tools perform only mono-core simulations and do not take advantage of the omnipresent multi-core processors. This paper addresses the problem of efficient parallelization of co-simulations. It presents a multi-core scheduling heuris-tic for parallelizing FMI-compliant models on multi-core processors. The limitations of this heuristic are highlighted and two solutions for dealing with them are presented. The obtained speed-up using each of these solutions is illustrated and discussed for further improvements .
Fichier principal
Vignette du fichier
confModelica2016.pdf (540.34 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01400033 , version 1 (21-11-2016)



Salah Eddine Saidi, Nicolas Pernet, Yves Sorel, Abir Ben Ben Khaled. Acceleration of FMU Co-Simulation On Multi-core Architectures. Japanese Modelica Conference, May 2016, Tokyo, Japan. pp.106 - 112, ⟨10.3384/ecp16124106⟩. ⟨hal-01400033⟩
395 View
388 Download



Gmail Facebook X LinkedIn More