WSC 2006 Abstracts

Compiled Code in Distributed Logic Simulation
Jun Wang and Carl Tropper (McGill University)

Abstract:
A logic simulation approach known as compiled-code event-driven simulation was developed in the past for sequential logic simulation. It improves simulation performance by reducing the logic evaluation and propagation time. In this paper we describe the application of this approach to distributed logic simulation. Our experimental results show that using compiled code can greatly improve the stability and overall performance of a Time-Warp based logic simulator. We also present a technique called fanout aggregation that makes use of information on circuit partitions and considerably improves the run-time performance of our (distributed) compiled code simulator. It does not produce a similar improvement when used in conjunction with an interpreted simulator because of run-time overhead.

Causality Information and Fossil Collection in Time Warp Simulations
Malolan Chetlur (AT&T) and Philip A Wilsey (University of Cincinnati)

Abstract:
This paper presents a Time Warp fossil collection mechanism that functions without need for a GVT estimation algorithm. Effectively each Logical Process (LP) collects causality information during normal event execution and then each LP utilizes this information to identify fossils. In this mechanism, LPs use constant size vectors (that are independent of the total number of parallel simulation objects) as timestamps called Plausible Total Clocks to disseminate causality information. For proper operation, this mechanism requires that the communication layer preserves a FIFO ordering on messages. A detailed description of this new fossil collection mechanism and its proof of correctness is presented in this paper.

Eliminating Remote Message Passing in Optimistic Simulation
David Bauer (The MITRE Corporation) and Christopher D Carothers (Rensselaer Polytechnic Institute)

Abstract:
This paper introduces an algorithm for parallel simulation capable of executing the critical path without a priori knowledge of the model being executed. This algorithm is founded on the observation that each initial event in a model causes a stream of events to be generated for execution. By focusing on the parallelization of event streams, rather than logical processes, we have created a new simulation engine optimized for large scale models (i.e., models with 1 million LPs or more).

Incremental Checkpointing with Application to Distributed Discrete Event Simulation
Thomas Huining Feng and Edward A. Lee (Center for Hybrid and Embedded Software Systems (CHESS))

Abstract:
Checkpointing is widely used in robust fault-tolerant applications. We present an efficient incremental checkpointing mechanism. It requires to record only the state changes and not the complete state. After the creation of a checkpoint, state changes are logged incrementally as records in memory, with which an application can spontaneously roll back later. This incrementalism allows us to implement checkpointing with high performance. Only small constant time is required for checkpoint creation and state recording. Rollback requires linear time in the number of recorded state changes, which is bounded by the number of state variables times the number of checkpoints. We implement a Java source transformer that automatically converts an existing application into a behavior-preserving one with checkpointing functionality. This transformation is application-independent and application-transparent. A wide range of applications can benefit from this technique. Currently, it has been used for distributed discrete event simulation using the Time Warp technique.

Performance Evaluation of a CMB Protocol
Célia Leiko Ogawa Kawabata (Centro Universitário Central Paulista), Regina Helena Carlucci Santana, Marcos José Santana, and Sarita Mazzini Bruschi (ICMC/USP) and Kalinka Regina Lucas Jaquie Castelo Branco (Centro Universitário Eurípedes de Marília)

Abstract:
This paper presents the performance evaluation of a CMB (Chandy-Misra-Bryant) protocol from the perspective of execution time. The performance of each logical process in simulation is measured. Our evaluation shows that logical processes can have different behaviors and different protocols can be used simultaneously in simulations. While some logical processes may perform well using conservative protocols, others can use optimistic protocols because otherwise most of the time these processes would be blocked unnecessarily. In order to analyze the behavior of the simulations some models were simulated using a CMB implementation called ParSMPLX. These models showed that each logical process of a simulation has a different behavior that makes it more suitable for a specific protocol, increasing the performance.

Efficient Parallel Queuing System Simulation
Tobias Kiesling (Universität der Bundeswehr München) and Thomas Krieger (Institut für Technik Intelligenter Systeme, Universität der Bundeswehr München )

Abstract:
Queuing systems are an important building block for performance evaluation in various application areas, due to their powerful, yet simple nature. Although it is often possible to perform an analytical evaluation of a queuing model, simulation of queuing systems remains an important technique in the context of performance evaluation. In order to speed up queuing simulation executions, parallel and distributed simulation techniques have been devised. Unfortunately, existing methods are complex in nature, leading to increased development costs. Moreover, most of these approaches have been developed for tightly coupled parallel processing machines. Consequently, they are not suited for a distributed computing environment. This paper investigates an alternative approach based on the technique of time-parallel simulation with fix-up computations. The salient features of this novel approach are its simplicity and its suitability for execution in a distributed environment.

A Distributed Simulation Approach for Modeling and Analyzing Systems of Systems
Abeer Tarief Sharawi, Serge N. Sala-Diakanda, Adam Dalton, Sergio Quijada, Nabeel Yousef, Jose Sepulveda, and Luis Rabelo (University of Central Florida)

Abstract:
Certain business objectives cannot be met without the interaction and communication between different systems. An interesting concept called system of systems (SoS), which aims to describe this interaction between systems has been gaining attention in the last few years. In this paper an extensive review of the literature is performed to capture the main characteristics associated to this concept in order to propose a new, more complete definition. This paper also proposes the use of distributed simulation through the High Level Architecture (HLA) rules to model and simulate systems of systems. We illustrate our idea with two different examples: a simplified supply chain network of a computer assembly and an aircraft initial sizing scenarios. The paper concludes with a discussion of some of the significant advantages distributed simulation could offer over traditional simulation for the analysis of such complex systems.

Development of a Runtime Infrastructure for Large-Scale Distributed Simulations
Buquan Liu, Yiping Yao, Jing Tao, and Huaiming Wang (School of Computer, National University of Defense Technology)

Abstract:
With the development of distributed modeling and simulation, it is necessary for the RTI to support large-scale applications. However, many RTIs can not support large-scale distributed simulations with more than 100 federates very well nowadays. StarLink+ is an RTI developed according to the IEEE 1516 standard, which can be used for large-scale simulations with thousands of federates. Great innovations are made in StarLink+, such as its architecture and inner implementation technologies. This paper presents the two-level architecture in StarLink+. The unique architecture has the advantages of both central architecture and distributed architecture. To improve the performance much more for large-scale simulations, two important technologies, i.e. multiple threads and data packing, are adopted in StarLink+. In addition, this paper explains the efficient advancing mechanism in time management and discusses the large-scale experiments with thousands of federates in StarLink+.

Implementation of Time Management in a Runtime Infrastructure
Buquan Liu, Yiping Yao, Jing Tao, and Huaimin Wang (School of Computer, National University of Defense Technology)

Abstract:
The High Level Architecture (HLA) time management is concerned with mechanisms for guaranteeing message order, process synchronization and execution correctness in distributed simulations. Time management greatly influences on the scales of applications, especially for the computation of Greatest Available Logical Time (GALT) and the implementation of optimistic services. StarLink is an RTI with central architecture, which is compliant with the IEEE 1516 standard. This paper systematically describes the implementation algorithms for main time management services in StarLink. Two smart and efficient algorithms about GALT computation and optimistic services are also introduced, which are suitable for many RTIs such as RTI1.3-NG, pRTI and DRTI. For the GALT algorithm, it is not necessary for an RTI to resolve the recursion nor any deadlock. For optimistic services, a simple mechanism without rollback in an RTI is also introduced; therefore, it can greatly simplify the development of an RTI.

Distributed Simulation in Industry – A Survey, Part 1 – the COTS Vendors
Csaba Attila Boer (TBA BV), Arie de Bruin (Delft University of Technology, Faculty of Electrical Engineering, Mathematics and Computer Science) and Alexander Verbraeck (Delft University of Technology, Faculty of Technology, Policy and Management)

Abstract:
Distributed simulation is used very little in industry, especially when compared with the interest in distributed simulation from research and from the military domain. In order to answer the question why industry lags behind, the authors have carried out an extensive survey, using a questionnaire and interviews, with users, vendors, and developers of distributed simulation products, as well as with vendors of non-distributed simulation software. Based on the results the discrepancies between the different “worlds” become clear enough to enable the formulation of clear guidelines for further developments of standards for distributed simulation. This paper reports on the first part of the survey, namely a questionnaire targeted at vendors of commercial-off-the-shelf (COTS) simulation packages. Analysis of the answers obtained establish that it is indeed the case that industry is relatively underdeveloped in the area of distributed simulation and also sheds some light on the reasons behind this.

Distributed Simulation in Industry – a Survey, Part 2 – Experts on Distributed Simulation
Csaba Attila Boer (TBA BV), Arie de Bruin (Delft University of Technology, Faculty of Electrical Engineering, Mathematics and Computer Science) and Alexander Verbraeck (Delft University of Technology, Faculty of Technology, Policy and Management)

Abstract:
Distributed simulation is used very little in industry, especially when compared with the interest in distributed simulation from research and from the military domain. In order to answer the question why industry lags behind, the authors have carried out an extensive survey, using a questionnaire and interviews, with users, vendors, and developers of distributed simulation products, as well as with vendors of non-distributed simulation software. This paper reports on the second part of the survey, namely a series of open ended interviews. We report on the responses we obtained indicating the discrepancies between the different “worlds”. A categorization of these responses is given using which it is possible to formulate clear guidelines for further developments of standards for distributed simulation.

Optimistic-conservative Synchronization in Distributed Factory Simulation
Leon McGinnis and Sheng Xu (Georgia Institute of Technology)

Abstract:
Distributed simulation is attractive for modeling complicated manufacturing systems having many tools and products, such as a semiconductor wafer fabrication line. However, conservative synchronization approaches can introduce excessive overhead in execution, and result in little parallelism, which can eliminate the speedup promised by distributed simulation. Our experiences in building a distributed simulation model for 300mm wafer fab using the High Level Architecture (HLA) shows that using model specific information in a novel adaptation of conservative synchronization can achieve very significant reduction in model execution time. This paper defines the time-chop problem for which this adaptation is effective, and formally develops our optimistic-conservative synchronization scheme.

Engineering ab Initio Dynamic Interoperability and Composability Via Agent-mediated Introspective Simulation
Levent Yilmaz (Auburn University) and Andreas Tolk (Virginia Modeling Analysis and Simulation Center)

Abstract:
Complex software intensive simulation systems must respond to changing technology, environments, and requirements. Hence, dynamic extensibility and adaptability is a significant concern in an application domain. However, existing interoperability and composability solutions are limited in dealing with dynamically evolving content needs of existing simulations and run-time inclusion of new components into a simulation system. Simulations that are dynamically extensible, while being interoperable require principled designs that facilitate engineering extensibility, interoperability, and composability in the first place. We propose and examine the utility of a strategy based on an agent-mediated meta-simulation architecture.

Composing Simulations From XML-Specified Model Components
Mathias Röhl and Adelinde M. Uhrmacher (University of Rostock)

Abstract:
This paper is about the flexible composition of efficient simulation models. It presents the realization of a component framework that can be added as an additional layer on top of simulation systems. It builds upon platform independent specifications of components in XML to evaluate dependency relationships and parameters during composition. The process of composition is split up into four stages. Starting from XML documents component instances are created. These can be customized and arranged to form a composition. Finally, a composition is transformed to an executable simulation model. The first three stages are general applicable to simulation systems; the last one depends on the Parallel DEVS formalism and the simulation system James II.

A Domain-specific Language for Model Coupling
Tom Bulatewicz and Janice Cuny (University of Oregon)

Abstract:
There is an increasing need for the comprehensive simulation of complex, dynamic, physical systems. Often such simulations are built by coupling existing, component models so that their concurrent simulations affect each other. The process of model coupling is, however, a non-trivial task that is not adequately supported by existing frameworks. To provide better support, we have developed an approach to model coupling that uses high level model interfaces called Potential Coupling Interfaces. In this work, we present a visual, domain-specific language for model coupling, called the Coupling Description Language, based on these interfaces. We show that it supports the resolution of model incompatibilities and allows for the fast-prototyping of coupled models.

Developing Interoperability Standards for Distributed Simulation and COTS Simulation Packages with the CSPI PDG
Simon J. E. Taylor (Centre for Applied Simulation Modelling), Stephen J Turner, Malcolm Yoke Hean Low, and Xiaoguang Wang (Nanyang Technological University), Steffen Strassburger (Fraunhofer Institute for Factory) and John Ladbrook (Ford Motor Company)

Abstract:
For many years discrete-event simulation has been used to analyze production and logistics problems in manufacturing and defense. In the early 1980s, visual interactive modelling environments were created that supported the development, experimentation and visualization of simulation models. Today these environments are termed Commercial-off-the-shelf Simulation Packages (CSPs). With the advent of distributed simulation and, later, the High Level Architecture, the possibility existed to link together these CSPs and their models to simulate larger problems within enterprises (e.g. multiple production lines) and across supply chains. However, the problem of standardizing the use of the HLA and its constituent parts in this domain exists. The solution of this problem is the work of the CSP Interoperability Product Development Group (CSPI PDG). The purpose of this paper is to introduce the CSPI PDG and to review the suite of standards proposed by the group and current progress.

The Road to COTS-Interoperability: From Generic HLA-Interfaces Towards Plug-and-Play Capabilities
Steffen Strassburger (Fraunhofer IFF)

Abstract:
Interoperability between commercial-off-the-shelf (COTS) simulation packages is a topic which has been discussed for many years without a solution. With the advent of the High Level Architecture for Modeling and Simulation (HLA) for the first time a real industry standard has been made available which promises interoperability for a wide range of simulation systems and applications. Successful attempts to integrate HLA interfaces into different simulation packages have been made in the past. However, these interfaces typically place a significant overhead on the simulation developer. Also, as often a generic HLA interface is provided, different HLA interfaces for different simulation packages are not necessarily interoperable per se, as there are different possible ways to use HLA for the same task. This article addresses these issues and discusses interoperability solutions based on and beyond of HLA. It further investigates the interoperability reference solutions put forward by the COTS simulation package integration forum.

About the Need for Distributed Simulation Technology for the Resolution of Real-world Manufacturing and Logistics Problems
Peter Lendermann (Singapore Institute of Manufacturing Technology)

Abstract:
Distributed simulation has undergone several cycles of ups and downs in recent years. Although successful in the military domain, it appears that the idea of applying distributed simulation in other fields for modeling and analysis of large-scale, heterogeneous systems such as communication networks or supply chains has still not taken off until today. Is this because of inherent limitations or lack of applicability as such? Or is it because of additional research issues that are yet to be resolved to make distributed simulation applicable? In this paper, the problem is discussed specifically with regard to the application of distributed simulation for design, operation and performance enhancement of manufacturing and logistics systems.

Interoperating Simulations of Automatic Material Handling Systems and Manufacturing Processes
Boon Ping Gan and Lai Peng Chan (Singapore Institute of Manufacturing Technology) and Stephen John Turner (Nanyang Technological University)

Abstract:
To perform a high fidelity simulation study on a 300 mm wafer fabrication plant, modeling of the manufacturing process (MP) alone is not sufficient. Inclusion of the automated material handling system (AMHS) model is necessary due to the high degree of factory automation. But there is no one tool that is capable of modeling both the AMHS and MP with sufficient accuracy and granularity. A commercial simulation package such as AutoMod is usually used to model the AMHS while AutoSched AP is usually used to model the MP. These packages can be integrated using the supplied interoperation module but flexibility in optimizing the execution performance for different simulation models is lacking. In this paper, we present an approach to interoperation based on the High Level Architecture standard. We note that the typical characteristics of disparity in the models’ time granularity and frequent model interactions are the obstacle to good execution performance.

Distributed Simulation with COTS Simulation Packages: A Case Study in Health Care Supply Chain Simulation
Navonil Mustafee and Simon J. E. Taylor (Brunel University) and Korina Katsaliaki and Sally Brailsford (University of Southampton)

Abstract:
The UK National Blood Service (NBS) is a public funded body that is responsible for distributing blood and associated products. A discrete-event simulation of the NBS supply chain in the Southampton area has been built using the commercial off-the-shelf simulation package (CSP) Simul8^TM. This models the relationship in the health care supply chain between the NBS Processing, Testing and Issuing (PTI) facility and its associated hospitals. However, as the number of hospitals increase simulation run time becomes inconveniently large. Using distributed simulation to try to solve this problem, researchers have used techniques informed by SISO's CSPI PDG to create a version of Simul8^TM compatible with the High Level Architecture (HLA). The NBS supply chain model was subsequently divided into several sub-models, each running in its own copy of Simul8^TM. Experimentation shows that this distributed version performs better than its standalone, conventional counterpart as the number of hospitals increases.

Reference Models for Supply Chain Design and Configuration
Markus Rabe, Frank-Walter Jaekel, and Heiko Weinaug (Fraunhofer IPK)

Abstract:
Today more and more essential processes are conducted across enterprise borders, inducing additional challenges in terms of different languages, process types and ontology. Business Process Modelling (BPM) and Simulation are well-understood methods to analyze and optimize the processes within an enterprise. However, they can also be used for cross-organizational application, especially if they are combined with reference structures. This paper explains techniques which support cross-enterprise design and configuration based on Reference Models. Thereby, different approaches such as SCOR, the Integrated Enterprise Modelling (IEM) and a specific Distributed Simulation Method are used and integrated into a consistent Reference Model approach. The application of this approach is illustrated with different projects which each focus on a specific aspect of the supply chain design and configuration.

Using Ontologies for Simulation Modeling
Perakath Benjamin (Knowledge Based Systems Inc.) and Mukul Patki, Michael Graul, Madhav Erraguntla, and Kumar Akella (Knowledge Based System Inc.)

Abstract:
Ontological analysis has been shown to be an effective first step in the construction of robust knowledge based systems. However, the modeling and simulation community has not taken advantage of the benefits of ontology management methods and tools. Moreover, the popularity of semantic technologies and the semantic web has provided several beneficial opportunities for the modeling and simulation communities of interest. This paper describes the role of ontologies in facilitating simulation modeling. It outlines the technical challenges in distributed simulation modeling and describes how ontology-based methods may be applied to address these challenges. The paper concludes by describing an ontology-based solution framework for simulation modeling and analysis and outlining the benefits of this solution approach.

An Ontology for Trajectory Simulation
Umut Durak (TUBITAK-SAGE), Halit Oguztuzun (Dept. of Computer Engineering, Middle East Technical University) and S. Kemal Ider (Dept. of Mechanical Engineering, Middle East Technical University)

Abstract:
From the concept exploration for a weapon system to training simulators, from hardware-in-the-loop simulators to mission planning tools, trajectory simulations are used throughout the life cycle of a weapon system. A trajectory simulation can be defined as a computational tool to calculate the flight path and flight parameters of munitions.There is a wide span of trajectory simulations differing widely with respect to their performance and fidelity characteristics, from simple point-mass simulations to sixseven degrees of freedom hardware-in-the-loop missile simulations. From our observations, it is a common practice in the industry that developments of these simulations are carried out as isolated projects although they rely on the same body of knowledge. We envision an ontology that will capture the common knowledge in trajectory simulation domain and make domain knowledge available for reuse. Trajectory Simulation Ontology, dubbed TSONT, is being developed to realize this vision.

Ontology Based Representations of Simulation Models Following the Process Interaction World View
Gregory A. Silver (University of Georgia), Lee W. Lacy (University of Central Florida) and John A. Miller (University of Georgia)

Abstract:
The Discrete Event Simulation (DES) process interaction world view describes models that focus on simulated entities that progress through a series of temporally related activities. DES formalisms and vendor approaches for representing DES models serve as a basis for developing an open neutral representation of models that can be encoded into ontologies. This paper reviews world views, formal foundations, and ontologies as background. The process for creating ontologies for the process interaction DES domain is discussed. An approach to ontology based simulation model representation is presented. Conclusions and recommendations for future work are provided.

On the Performance of Inter-Organizational Design Optimization Systems
Paolo Vercesi (Esteco) and Alberto Bartoli (DEEI)

Abstract:
Simulation-based design optimization is a key technology in many industrial sectors. Recent developments in software technology have opened a novel range of possibilities in this area. It has now become possible to involve multiple organizations in the simulation of a candidate design, by composing their respective simulation modules on the Internet. Thus, it is possible to deploy an inter-organizational design optimization system, which may be particularly appealing because modern engineering products are assembled out of smaller blocks developed by different organizations. In this paper we explore some of the fundamental performance-related issues involved in such a novel scenario, by analyzing a variety of options: centralized control vs. distributed control; generation of new candidate designs one at a time or in batches; communication and computation performed serially or with time overlap. Our analysis provides useful insights into the numerous trade-offs involved in the implementation of inter-organizational design optimization.

Using Java Methods Traces to Automatically Characterize and Model J2EE Server Applications
Darpan Dinker and Herb Schwetman (Sun Microsystems, Inc.)

Abstract:
This paper describes a novel framework used to characterize a J2EE (Java Enterprise Edition) application and develop models of the application by using Java method tracing in a Java-technology based application server. Application servers are critical to large-scale, online servers and serve as middleware to provide secure access to transactional, legacy and web services. The tracing tool in this framework gives a detailed and comprehensive view of the sequences of methods invoked as the application server processes requests. The output of this tool is processed and automatically summarized into a set of transaction profiles which form the input for a simulation model of the application server and its related components. These profiles have proven to be a useful abstraction of the behavior of the transactions processed by the system. After describing the tool and the model, the paper provides results of validation runs and applications of these techniques.

Simulation of Job Scheduling for Small Scale Clusters
Hassan Rajaei, Mohammad B Dadfar, and Pankaj Joshi (Bowling Green State University)

Abstract:
Despite the growing popularity of small-scale clusters built out of off-the-shelf components, there has been little research on how these small-scale clusters behave under different scheduling policies. Batch scheduling policies with backfilling provide excellent space-sharing strategy for parallel jobs. However, as the performances of uniprocessor and symmetric multiprocessor have improved with time-sharing scheduling strategies, it is intuitive that the performance of a cluster of PCs with distributed memory may also improve with time-sharing strategies, or a combination of time-sharing and space-sharing strategies. Apart from the batch scheduling policies, this research explores the possibilities of using synchronized time-sharing scheduling algorithms for clusters. This paper describes simulation of the Gang scheduling policies on top of an existing batch scheme. The simulation results indicate that time-sharing scheduler for clusters could exhibit superior performance over a batch policy.