Research Article
Experimental Comparison of Open Source Discrete-Event Simulation Frameworks
@INPROCEEDINGS{10.1007/978-3-030-97124-3_24, author={Oskar Skak Kristiansen and Ulrik Sandberg and Casper Hansen and Morten Skovgaard Jensen and Jonas Friederich and Sanja Lazarova-Molnar}, title={Experimental Comparison of Open Source Discrete-Event Simulation Frameworks}, proceedings={Simulation Tools and Techniques. 13th EAI International Conference, SIMUtools 2021, Virtual Event, November 5-6, 2021, Proceedings}, proceedings_a={SIMUTOOLS}, year={2022}, month={3}, keywords={Discrete-event simulation Open source Performance comparison}, doi={10.1007/978-3-030-97124-3_24} }- Oskar Skak Kristiansen
Ulrik Sandberg
Casper Hansen
Morten Skovgaard Jensen
Jonas Friederich
Sanja Lazarova-Molnar
Year: 2022
Experimental Comparison of Open Source Discrete-Event Simulation Frameworks
SIMUTOOLS
Springer
DOI: 10.1007/978-3-030-97124-3_24
Abstract
Simulation is a growing and very relevant field today. With many application domains and increasing computational power, simulations are solving an increasing number of real-world problems in a safe and efficient manner. The aim of this study is to make an experimental comparison among popular open-source discrete-event simulation frameworks, selected through a thorough survey. The experimental comparison considers numerous parameters, the most important of which were studied via an operational experiment of performance, as well as a survey of surface level characteristics and usability. We initially surveyed 50 frameworks for high-level parameters relating to corresponding implementations, filtered by the generality of domain, license, popularity, etc. As a result, we shortlisted five frameworks, which we then performance-tested in an operational experiment. We discovered significant performance differences under the given loads. We also completed a usability survey to provide a holistic impression of the frameworks, further identifying key differences.
