Scalability analysis of cloud-based distributed
simulations of IoT systems using HLA
Thomas Nägele and Jozef Hooman
Proceedings 24th International Conference on Parallel and Distributed Systems (ICPADS 2018), IEEE, pages 1075-1080, 2018.
Gaining insight in the properties of an Internet of
Things (IoT) system during the design phase is difficult.
The cosimulation of such a system would be very useful, but creating it
is usually time consuming. By means of domain specific languages
(DSLs) we support the fast construction of large co-simulations
of IoT systems. This approach includes the use of CoHLA, a DSL
that generates co-simulation code based on the HLA and FMI
standards. Due to the large number of connected sensors and
actuators in an IoT system, the time needed for simulation can
be a blocking factor. Hence we facilitate distributed co-simulation
in the cloud. To do that efficiently, we have conducted a set of
experiments to analyse scalability and the performance impact
of distribution methods. From these experiments, lessons were
learned on how to distribute the co-simulation of IoT systems.