The Community for Technology Leaders
2012 ACM/IEEE/SCS 26th Workshop on Principles of Advanced and Distributed Simulation (2003)
San Diego, California
June 10, 2003 to June 13, 2003
ISSN: 1087-4097
ISBN: 0-7695-1970-9
pp: 40
Lorenzo Donatiello , Dipartimento di Scienze dell Informazione, Universit? degli Studi di Bologna,
Gabriele D'Angelo , Dipartimento di Scienze dell?Informazione, Universit? degli Studi di Bologna,
Luciano Bononi , Dipartimento di Scienze dell?Informazione, Universit? degli Studi di Bologna,
ABSTRACT
Wireless networks? models differ from wired ones at least in the innovative dynamic effects of host-mobility and open-broadcast nature of the wireless medium. Topology changes due to simulated hosts? mobility map on causality effects in the "areas of influence" of each mobile device. The analysis of wireless networks of interest today may include a potentially high number of simulated hosts, resulting in performance and scalability problems for discrete-event sequential simulation tools and methods, on a single physical execution unit (PEU). In a distributed simulation, the main bottleneck becomes the communication and synchronization required to maintain the causality constrains between distributed model components. In this work we propose a HLAbased, dynamic mechanism for the runtime management and allocation of model entities in a distributed simulation of wireless networks models, over a cluster of PEUs. By adopting a runtime evaluation of causal bindings between model entities we map the causal effects of virtual topology changes to dynamic migration of data structures. Preliminary results demonstrate that the prototype heuristics lead to a reduction in the percentage of external communication between the PEUs, limited overheads and performance enhancements for a worstcase scenario.
INDEX TERMS
null
CITATION
Lorenzo Donatiello, Gabriele D'Angelo, Luciano Bononi, "HLA-based Adaptive Distributed Simulation of Wireless Mobile Systems", 2012 ACM/IEEE/SCS 26th Workshop on Principles of Advanced and Distributed Simulation, vol. 00, no. , pp. 40, 2003, doi:10.1109/PADS.2003.1207419
100 ms
(Ver 3.3 (11022016))