WSC 2008 Final Abstracts

Supporting Simulation in Industry Through the Application of Grid Computing
Navonil Mustafee (Warwick Business School) and Simon J. E. Taylor (Brunel University)

Abstract:
An increased need for collaborative research, together with continuing advances in communication technology and computer hardware, has facilitated the development of distributed systems that can provide users access to geographically dispersed computing resources that are administered in multiple computer domains. The term grid computing, or grids, is popularly used to refer to such distributed systems. Simulation is characterized by the need to run multiple sets of computationally intensive experiments. Large scale scientific simulations have traditionally been the primary benefactor of grid computing. The application of this technology to simulation in industry has, however, been negligible. This research investigates how grid technology can be effectively exploited by users to model simulations in industry. It introduces our desktop grid, WinGrid, and presents a case study conducted at a leading European investment bank. Results indicate that grid computing does indeed hold promise for simulation in industry.

Management of HLA-Based Distributed Legacy SLX-Models
Thomas Schulze (University of Magdeburg), Steffen Strassburger (Technical University of Ilmenau) and Michael Raab (Fraunhofer IFF Magdeburg)

Abstract:
This article discusses management tasks that have to be supported for an efficient reuse of simulation models within HLA-based distributed simulation environments. After a review of the controversial discussions on the deployment of distributed simulation in the industrial domain, this article derives the need of support for such basic management tasks. Based on the practical experiences from an industrial project from the manufacturing domain, in which several legacy SLX models had to be integrated, we demonstrate how these management tasks have been supported and which tools were needed for this purpose. We furthermore discuss the results of the distributed simulation and show the necessity and the added benefit provided through its usage in industry.

Distributed Simulation in Industry – A Survey, Part 3 – The HLA Standard in Industry
Csaba A. Boer (TBA B.V.) and Arie de Bruin and Alexander Verbraeck (Delft University of Technology)

Abstract:
Distributed simulation, more specifically the HLA standard, is hardly applied in industry. We have conducted an extensive survey with COTS (commercial off-the-shelf) simulation package vendors and simulation experts, both from defence and industry, that focuses, amongst others, on the question what the reasons are behind this phenomenon. In this paper we analyze the reactions that we obtained, categorizing them into arguments related to distributed simulation in general, arguments related to HLA and arguments pertaining to the embedding of HLA concepts in COTS packages. These answers will lead us, we believe, to insights that can serve as guidelines to make distributed simulation more attractive for the industrial simulation community.

Predictive-Conservative Synchronization for Commercial Simulation Package Interoperability
Yuanxi Liang and Stephen John Turner (Nanyang Technological University) and Boon Ping Gan (D-SIMLAB Technologies Pte Ltd)

Abstract:
Distributed simulation is desired in many industries to support analysis and decision making for complex and integrated problems. Interoperating commercial simulation packages using the High Level Architecture (HLA) is a technique to fulfill this demand. Time management is an important factor in determining the execution performance when using this technique. In some simulation models, conservative synchronization does not provide good efficiency because of the high cost of requesting a time advance with every event. In this paper, we describe a user-friendly framework for interoperating AutoMod models by adopting the HLA standard with a new time management mechanism. A "departure port" mechanism identifies the characteristics of the message outgoing process and provides a dynamic safe request time to invoke the time advancing service. We conducted experiments using close-to-reality models which show that the execution time can be reduced by over 50%.

Improving Performance by Replicating Simulations with Alternative Synchronization Approaches
Zengxiang Li, Wentong Cai, Stephen John Turner, and Ke Pan (Nanyang Technological University)

Abstract:
Parallel and distributed simulation facilitates the construction of a simulation application (i.e., federation in HLA terminology) with a number of simulation components (federates). Recently, an approach based on active replication technique has been proposed to improve the performance of simulations by exploring software diversity. To guarantee the correctness of the approach, all replicas of the same federate are required to be Piece-Wise-Deterministic (PWD). However, the PWD restriction will not be satisfied if the replicas achieve software diversity by employing different kinds of synchronization approaches. A replication structure is proposed in this paper, which can transparently handle the federate replicas that use either conservative or optimistic synchronization approach. The execution speed of the federation is increased by always choosing the fastest replica to represent the federate in the federation. Besides presenting the implementation details, we also report the experimental results to demonstrate the performance improvement of the replication structure.

Federated Simulations for Systems of Systems Integration
Robert Kewley, Edward Teague, and Dale Henderson (USMA), Niki Goergor (US Army Corps of Engineers Engineering Research and Devlopment Center) and James Cook (USMA)

Abstract:
Systems of systems integration is a difficult engineering challenge that places a particular burden on the engineers who must develop simulation models to support that integration. Developing a large scale stand-alone model to support systems integration is a time-consuming process that is often not possible. An alternative approach is to leverage existing models in a federation. This type of work requires a specialized set of engineering skills. The United States Military Academy Department of Systems Engineering SysHub research program is better defining these skills and applying them to different problem domains. This paper highlights how capabilities for information exchange, environmental representation, entity representation, model development, and data collection support the federation development process.

Design and Implementation of an XML-Based, Technology-Unified Data Pipeline for Interactive Simulation
Francois Rioux (Laval University), Francois Bernier (DRDC-Valcartier) and Denis Laurendeau (Laval University)

Abstract:
Providing software that is efficient, flexible, reusable and easy to work with is a hard task for simulation developers. In this paper we propose the use of XML and its related tools (e.g. JAXB, XQuery, XSLT, and Native XML Database) for the implementation of a technology-unified data pipeline targeted to interactive simulation. We introduce a technology-independent conceptual data model as the basis for every simulation framework. We show that XML is a well-suited technology to be used in that context. We propose a data modeling methodology that takes its roots from Model-Driven Engineering (MDE). We also show a sample implementation that uses XML for transmitting data over the entire simulation loop. We thus present our experience in implementing that kind of architecture and discuss how the use of XML and associated technologies help in building a unified and generic data pipeline for interactive simulation.

Service-Oriented-Architecture Based Framework for Multi-User Virtual Environments
Xiaoyu Zhang and Denis Gracanin (Virginia Tech)

Abstract:
Service-Oriented Architecture (SOA) is an application framework used for creating complex enterprise systems by integrating distributed services. The SOA standards are primarily focused on the service composability and data interoperability. However, SOA has its limitations in terms of the performance of real-time message exchanging. To address the disadvantages and improve the application performance, we propose a framework that combines the streaming technology and SOA. The framework is used for constructing multi-user Virtual Environment applications by integrating the application content from distributed services. The additional streaming channels applied to SOA enable the services to actively propagate the real-time messages. Due to the distributed architecture of SOA and the heavy payload of message exchange, the application performance needs to be evaluated. We describe the metrics used to evaluate the performance and present the evaluation results. Based on the experiment results, we discuss the categories of applications that fit well in our framework.

Knowledge Representation and the Dimensions of a Multi-Model Relationship
Charles Daniel Turnitsa (Virginia Modeling Analysis and Simulation Center) and Andreas Tolk (Old Dominion University)

Abstract:
The exchange of data between different models in a multi-simulation environment is about the exchange of information from within the context of two separate world views. This amounts to knowledge (information in context). To represent this knowledge, in any of a number of different ways (meta-data, ontological reference, frames, etc.) there must be a method to bridge the gap between what the world view of origination can represent and what the world view of destination can receive. This requires some understanding of the dimensions of difference between the two world views. As each separate model is the product of a different perspective by the model’s developer, each individual model will have a different world view. Understanding what these differences are, and viewing them in light of the requirements for conceptual interoperability will contribute a great teal to the knowledge that must be captured for meaningful exchange.