Authors:
Moritz Gütlein
and
Anatoli Djanatliev
Affiliation:
Computer Networks and Communication Systems, Friedrich-Alexander University Erlangen-Nürnberg, Martensstr. 3, Erlangen, Germany
Keyword(s):
Modeling and Simulation as a Service, MSaaS, SaaS, Kafka, Docker, Simulation, Traffic Simulation, Virtual Mobility World.
Abstract:
Among other requirements, in the field of Modeling and Simulation there is a need to build and run (co-) simulation models on-demand. Setting up, orchestrating and executing simulations should be easy, while additional features such as interaction interfaces open up a variety of applications. Being able to connect external components seamlessly to a running simulation allows for elaborate experiments. This paper describes the architecture and design of a simulation platform that is a part of a broader platform enabling evaluations of future mobility scenarios. Apache’s Kafka platform for big data stream processing is used as a communication base in order to enable all these requirements, as well as for the coupling of different simulation tools forming a co-simulation. We give an overview of existing works regarding Modeling and Simulation as a Service, before explaining our own approach. Therefore, the architecture, the interfaces, and the workflow of a simulation run is described.
The approach is illustrated by a case study, which is used to measure the service’s overhead.
(More)