No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Parallel schedule method of SMP2 models under SOA
Abstract
To achieve the parallel schedule of SMP2 models under the Service Oriented Architecture, a parallel schedule method of SMP2 model is put forward. The assignment method of serial model schedule task is promoted based on the analysis of the four traditional serial model schedule tasks in the SMP2.Then three schedule tasks which can not be scheduled parallelized were reformed and the schedule processes were introduced. Finally, the schedule service interface was extended and the parallel simulation component interface, time manager interface and message manager interface were defined to conduct the implementation of parallel simulation component.
ResearchGate has not been able to resolve any citations for this publication.
For the development of large complex simulation systems, there are four basic model specification requirements, i.e. domain independence, platform independence, model architecture description support, simulation development and application process framework. Traditional simulation model representation specification methods fall into five categories, i.e. programming languages, simulation languages, modeling formalisms, generic simulation environments, and simulation interoperability protocols. Their shortages in meeting those four requirements are figured out. The applicability of Simulation Model Portability Standards 2 (SMP2) in representation complex simulation systems is detailed. The SMP2's advantages compared with prevalent High Level Architecture (HLA) and Base Object Model (BOM) specification are summarized. According to our four years research and more than two years application practice of SMP2, nine key application techniques for SMP2 are distilled and discussed. Among them several have been implemented and used in practical projects, the solutions of the rest are presented as the guides of next work.
Today, M&S combined with High Performance Computing is becoming the third important means for recognizing and rebuilding the objective world besides the theory research and experiment research. At present, the tendency of system modeling and simulation technology is developing towards networkitization, virtualization, intelligentization, collaboratization and pervasivization. Based on the research fruits of paper 1 and author's recent research projects, this paper further discusses the meanings of modern M&S and its architecture. Furthermore, some focusing points in recent research and application of M&S are also discussed and prospected, including networkitized M&S technology based on modern network techniques, M&S technology on synthetized nature environment, Intelligent system modeling and intelligent simulation system, M&S technology of complex system and open, complex, huge system, virtual prototyping engineering technology, high performance computer, pervasive simulation technology etc.. At last, some conclusions are given.
With the increasing requirements of complexity and scalability, simulation interoperability and composability become the most competitive research topics in field of M&S. We found that, by employing service-oriented architecture, Web service and grid technologies have provided large-scale distributed resource sharing and application coordinated developing. Motivated by the above observation, in this paper, we propose service-oriented dynamic composition architecture for distributed simulation, SOA4DSDC. This framework can provide component-based model-driven view for simulation modeling, and service-oriented view (i.e. loose coupling) for simulation execution. This paper studies the motivation for developing such a framework and presents the details of the development.
To satisfy requirements of complex system modeling and simulation, an M&S approach of combining SMP2 and SOA is improved. Aiming to implement SMP2-based simulation under SOA, four aspects of SMP2 are studied: 1) service-oriented distributed simulation system architecture is propounded. 2) Component interfaces related to the model composition and schedule are redefined. 3) Simulation model service description is added to the Simulation Modeling Define Language which is called SMP2 metamodel. 4) Modeling and simulation process for SMP2 under SOA is detailed. Finally, a SOA-based modeling and simulation environment for SMP2 is designed and implemented, and an example of Navigation System's Service Performance analysis simulation is given to demonstrate the validity and feasibility of the method.
Research on SOA-based Distributed SMP2 Parallel Simulation Architeture
- Wang Chao
- Song Lili
- Li Qun
- Wang Weiping