Actor-Based Simulation of PDEVS systems over HLA
Title | Actor-Based Simulation of PDEVS systems over HLA |
Publication Type | Conference Paper |
Year of Publication | 2008 |
Authors | Cicirelli, F, Furfaro, A, Nigro, L |
Conference Name | Simulation Symposium, 2008. ANSS 2008. 41st Annual |
Pagination | 229-236 |
Date Published | April |
Keywords | actor-based simulation, ActorDEVS/Theatre, Actors, Adaptive systems, agents, asynchronous message-passing, Biological system modeling, Communication system control, Context modeling, Control systems, digital simulation, Discrete event simulation, Engines, formal specification, HLA, Java, Laboratories, location-transparent naming, message passing, mobile agents, Modelling and simulation, network communications, parallel discrete event system specification, parallel simulation engine, PDEVS, PDEVS modelling, PDEVS simulation control engine, PDEVS systems, software architecture, tie-breaking mechanism, time management, variable structure systems |
Abstract | The work described in this paper aims at experimenting with PDEVS (parallel discrete event system specification) modelling and simulation of complex and variable structure systems. The approach is centred on the use of actors (agents), i.e. lightweight reactive objects which communicate to one another by asynchronous message-passing. Actors can migrate from a computing node to another. More specifically, this paper proposes a mapping of ActorDEVS, a lean control-based framework supporting PDEVS in Java, over HLA using an adaptation of the Theatre actor-based software architecture, which provides location-transparent naming and efficient network communications. The combination ActorDEVS/Theatre makes it possible to customize the PDEVS simulation control engine. The paper shows a parallel simulation engine with a time management compliant with HLA, which embodies a tie-breaking mechanism for simultaneous events. The practical use of the resultant approach is demonstrated through an example. |
DOI | 10.1109/ANSS-41.2008.5 |