Abstract: Programmable graphics processing units (GPUs) have emerged as excellent computational platforms for certain general-purpose applications. The data parallel execution capabilities of GPUs specifically point to the potential for effective use in simulations of agent-based models (ABM). In this paper, the computational efficiency of ABM simulation on GPUs is evaluated on representative ABM benchmarks. The runtime speed of GPU-based models is compared to that of traditional CPU-based implementation, and also to that of equivalent models in traditional ABM toolkits (Repast and NetLogo). As expected, it is observed that, GPU-based ABM execution affords excellent speedup on simple models, with better speedup on models exhibiting good locality and fair amount of computation per memory element. Execution is two to three orders of magnitude faster with a GPU than with leading ABM toolkits, but at the cost of decrease in modularity, ease of programmability and reusability. At a more fundamental level, however, the data parallel paradigm is found to be somewhat at odds with traditional model-specification approaches for ABM. Effective use of data parallel execution, in general, seems to require resolution of modeling and execution challenges. Some of the challenges are identified and related solution approaches are described.

Abstract: An important goal of software security is to ensure sensitive/secret data owned by a program shall be exclusively accessible by the program. An obstacle to such security goal is that modern commodity operating systems (OS) for the sake of speed and flexibility have a unified linear address space--any OS kernel program can access all the linear addresses. As a result, rootkits or malicious system software are able to control the OS virtual address space, harvest the sensitive data used by software programs on the compromised computer, and report the data to remote entities controlled by hackers.In this paper, we present a holistic approach against sophisticated malware. Instead of focusing on the security of various abstraction layers of OS, we utilize the hardware techniques to directly provide the trust services to software programs. Without modifying OS, we leverage the virtual machine monitor technologies to create a lightweight hypervisor for fine-grain software runtime memory protection. As a result, a program's memory could be hidden from other high privilege system software in a single commodity OS. In addition, we propose the data locker component in the hypervisor, which prevents the sensitive data of software program in persistent storage from leaking to rootkits or other malware. For the performance evaluation, the implementation based on hardware-assisted x86 virtualization technology is presented and experimental results are reported.

Abstract: Limited stockpiles of antiviral medications and lack of availability of early strain-specific vaccine will require a multi-component strategy of pharmaceutical and non-pharmaceutical measures to delay or contain a future catastrophic avian influenza pandemic. A strategy composed of the antiviral stockpile distribution, school closures, and social distancing, followed by strain-specific vaccine when available is proposed. The EpiSimS agent-based simulation model with a structured population is used to assess the effectiveness of this strategy and to explore the sensitivity of its elements, in particular the level of school closures and the start time for non-pharmaceutical interventions, with varying amounts of expected fear-based isolation behavior. Results show that the level of school closure has the largest effect on reducing morbidity and mortality, comparable to US seasonal flu when starting early with a high level of school closures. Small variations of fear-based isolation showed little impact on morbidity and mortality, though a large second-order effect is seen on worker absenteeism.

Abstract: On behalf of the Organizing Committee we welcome you to the 2008 Spring Simulation Multiconference (SpringSim'08), sponsored by The Society for Modeling and Simulation International (SCS) in collaboration with ACM/SIGSIM. SpringSim'08 brings together ten Symposia, providing a forum for academia, industry, business, military, and government. It covers a wide variety of disciplines and domains that exploit Modeling and Simulation (M&S) to present their work in a unique setting. SpringSim'08 has several new events. The conference is located for the first time in Canada, in the city of Ottawa. Canada's National Capital is a center of innovative high technology companies, Universities, research laboratories, and Government that recognize the importance of M&S in our society. The Canadian organizers have very active pursuing this occurrence and contributed significantly to the program. Second, a tutorial track is launched this year furnishing cutting edge and state-of-the-art technologies in M&S for the participants. A poster session for Ph.D. students is arranged for the Ph.D. candidates to share their latest findings with the conference audience. Modeling and Simulation in Education (MSE) is a new track devoted to current trends in training in the field. Last but not least, we are delighted to welcome for the first time in SpringSim the 17th Annual International Conference on Health Science Simulation (ICSHSS). The SpringSim'08 program includes a wide selection of technical presentations and distinguished speakers. Professionals, Engineers, and Scientists who are committed, involved, or interested in M&S will find in this year's conference a world-class collection of state-of-the-art presentations and articles related to research, development, and applications of M&S.

Abstract: Service-oriented Architectures (SOA) have become increasingly popular as a way to support the business processes of an organization.[1] A Service-Oriented Architecture approach to system design is one where application design and development are done with the goal of producing usable services. This approach allows for the integration of applications as reusable system services. The services must have platform independent specifications which abstract the underlying complexity of the service, are loosely coupled, and, perhaps most importantly, are reusable. This paper gives an in depth overview of SOA concepts, and will briefly introduce current Enterprise Architectures, such as the Department of Defense Architecture Framework (DoDAF), Ministry of Defense Architecture Framework (MoDAF), and The Open Group Architecture Framework (TOGAF), and what is being done to address the current need for SOA.

Abstract: A model is presented of the way that our cultural attitude towards the unknown influences the decisions we make in trade. Uncertainty avoidance is one of Hofstede's five cultural dimensions. The paper presents a model of how this dimension affects trade. This influence has been explicated for the decisions regarding trade: partner selection, negotiation behavior, trust, and the interpretation of the trade partner's behavior. It has been verified in simulations showing that the generic tendencies as attributed to uncertainty avoidance are reflected in the simulation results. Our approach is an example of instantiating generic knowledge on the influences of culture on decision-making in general.

Abstract: This paper firstly seeks an answer to the question of whether Non Player Characters (NPCs) in Computer Games can also be viewed as game agents where reactivity, autonomy, being temporarily continuous and having goal-oriented behavior are taken as the features of being a game agent. Within this scope, we will try to assess whether naming NPCs as agents point to a desire that one day they would fulfill the requirements of being an agent or whether NPCs actually already fulfill these requirements. Secondly, the paper looks into the AI needs of video games. We present the AI methodologies that are either being used and under research for use in Game AI. The same methodologies are also likely to contribute to the increasing levels of autonomy and goal-directed behavior of NPCs and help them become more agent-like.

Abstract: The Deep Green concept is an innovative approach to using simulation to support ongoing military operations while they are being conducted. The basic approach is to maintain a state space graph of possible future states. Software agents use information on the trajectory of the ongoing operation, vice a priori staff estimates as to how the battle might unfold, as well as simulation technologies, to assess the likelihood of reaching some set of possible future states. The likelihood, utility, and flexibility of possible future nodes in the state space graph are computed and evaluated to focus the planning efforts. This notion is called anticipatory planning and involves the generation of options (either automated or semi-automated) ahead of "real time," before the options are needed. In addition, the Deep Green concept provides mechanisms for adaptive execution, which can be described as "late binding," or choosing a branch in the state space graph at the last moment to maintain flexibility. By using information acquired from the ongoing operation, rather than assumptions made during the planning phase, commanders and staffs can make more informed choices and focus on building options for futures that are becoming more likely. This paper will describe the Deep Green concept in detail.

Abstract: Belief-Desire-Intention (BDI) is a powerful agent paradigm that allows for the development of so-called intelligent agents - agents that can reason and act based on their beliefs and intentions. However, this power often comes at the cost of increased computational overhead. We describe our experience using a BDI agent framework for developing a simulation of collaborative air traffic flow management and the efficiency problems we encountered. By using BDI more judiciously in our simulation, we were able to address these issues and greatly reduce the execution time of our simulation. From our successes and failures, we derive several guidelines that may enable other researchers to avoid similar efficiency issues in BDI-based simulations.

Abstract: This paper deals with the development of intelligent agents with respect to their process specifications The development process can be handled and documented by the standard UML tool UML activity diagrams were extended for our purposes - Agent Behaviour Diagram Next, the behaviour reconfiguration principle is described in more detail A learning mechanism of agents was specified thanks to mentioned reconfiguration principle Our approach was implemented in the area of traffic simulation

Abstract: Any organization that moves cargo and people needs to decide how many platforms it needs to accomplish its varied tasks In this paper, we develop the Stochastic Fleet Estimation or SaFE Model which estimates the size and composition of the average required fleet of vehicles or platforms The SaFE model is a Monte Carlo-based simulation that uses task frequency, duration and task-assigned platform information to determine what fleet can accomplish these tasks over the given time period Results based on a simulated data set are presented and analysed

Abstract: A read-only transaction (ROT) does not modify any data. The main issues regarding processing of read-only transactions (ROTs) are correctness, data currency and performance. Even though the popular two-phase locking protocol processes ROTs correctly with no data currency related issues, its performance deteriorates with data contention. To improve the performance of ROTs, snapshot isolation-based approaches have been proposed. Even though snapshot isolation-based approaches improve the performance of ROTs, both data currency of ROTs and correctness (serializability) are compromised. In this paper, we propose an asynchronous speculative locking protocol (called as ASLR) which improves the performance of ROTs by trading extra processing resources. The simulation results show that ASLR improves the performance of ROTs significantly over two-phase locking and snapshot isolation-based approaches with manageable extra processing resources. The ASLR approach processes ROTs without any data currency and correctness issues.

Abstract: Fundamentals of e-learning/e-teaching environments are reviewed Role of personality and emotions in e-learning are explained Learning styles and Myers-Briggs Type Indicator (MBTI) are explained Basics of personality and emotional filters in agent simulation of e-learning/e-teaching are covered In addition, in order to help the process of learning, the agent of virtual classmate is used in some levels of learning

Abstract: The problem of how the explicit and implicit assumptions made during agent and simulation development are formulated and handled (or not handled) has received comparatively little research. Effective cooperation between multiple artificial agents requires not only an explicit representation of terms, concepts and processes, but also alignment of meaning between developers, integrators, and users. Assumptions, especially implicit ones, have potentially tremendous impact on establishing unambiguous representation due to potentially unintended interpretations.This problem is important to agent directed simulation in several ways: agent mediated selection of simulation components requires methods for ensuring valid component interoperation; agents are a logical choice for automatically comparing sets of assumptions; and when potential conflicts are detected, agents have the potential to adjudicate and resolve them. This discussion will focus on the role of assumptions in modeling because it is fundamental to composing valid models and systems.This paper summarizes how assumptions are defined, characterized, used and misused by modelers. It examines how assumptions can be used to identify potential conflicts between domain views.

Abstract: Data Distribution Management (DDM) is one of the services provided by the High Level Architecture (HLA) Run-time Infrastructure (RTI), which aims to reduce the amount of irrelevant data communication among federates and to minimize computational requirements for sending and receiving messages.Several DDM methods, including region-based, grid-based and their hybrids have been proposed in the literature. In this paper, we propose a new quadtree-based approach which significantly reduces irrelevant data communication and system complexity compared to previous methods. We introduce a mini-RTI toolkit to evaluate our model and previous methods. The experimental results show the improvement of our approach compared to grid and region based methods.

Abstract: This paper is an attempt to explain and understand what security architecture means and represents. A starting point was to include all elements of security architecture such as: network, host-based, applications, information, software, hardware, databases and physical elements. Any security architecture should also include principles and process laid out in this paper. Models to capture security architecture and an example are presented. Finally techniques used to capture and assess security architectures are mentioned.

Abstract: Modeling and simulating train behaviours in large scale geographic environments is a complex process. Such a process involves several heterogeneous actors evolving and interacting with their environment. The use of statistical and mathematical models does not satisfy the requirements of such a complex process where spatial data is of fundamental importance. On the other hand, the Agent-Based Geo-Simulation provides a flexible approach that can be used to easily simulate complex systems in large scale georeferenced environments. In this paper we present Train-MAGS, an agent-based geosimulation tool which simulates train behaviours and identifies risky areas in large scale geographic environments. We describe the way in which using the agent based simulation opens perspectives regarding the development of new functionalities to improve the risk assessment of railway networks.

Abstract: We introduce an extension of the classic Discrete Event System Specification (DEVS) formalism that includes stochastic features. Based on the use of Probability Spaces, the STochastic DEVS specification (STDEVS) provides a formal framework for modeling and simulation of general non deterministic discrete event systems. The main theoretical properties of STDEVS are shown. We illustrate its use in a stochastic-oriented simulation example with the main purpose of performance analysis in computer systems and data networks.

Abstract: We describe procedures for converting a macromolecular regulatory model from the most common deterministic formulation to one suitable for stochastic simulation. To avoid error, we seek to automate as much of the process as possible. However, deterministic models often omit key information necessary to a stochastic formulation. In this paper we introduce how we implement conversion in the JigCell modeling environment. Our tool makes it easier for the modeler to include complete details. Stochastic simulations are known for being computationally intensive, and thus require high performance computing facilities to be practical. We provide the first stochastic simulation results for realistic cell cycle models, using Virginia Tech's System X supercomputer.

Abstract: Wikipedia, a User Innovation Community (UIC), is becoming increasingly influential source of knowledge. The knowledge in Wikipedia is produced and processed collaboratively by UIC. The results of this collaboration process present various seemingly complex patterns demonstrated by update history of different articles in Wikipedia. Agent simulation is a powerful method that is used to study the behaviors of complex systems of interacting and autonomous agents. In this paper, we study the collaborative knowledge processing in Wikipedia using a simple agent-based model. The proposed model considers factors including knowledge distribution among agents, number of agents, behavior of agents and vandalism. We use this model to explain content growth rate, number and frequency of updates, edit war and vandalism in Wikipedia articles. The results demonstrate that the model captures the important empirical aspects in collaborative knowledge processing in Wikipedia.

Abstract: This paper presents a method for view-dependent texture management into a large terrain 3D visualization system. The proposed system can handle several collections of images, each one subdivided in tiles and stored in a database management system (DBMS). The main contribution of this work is a methodology for managing multi-resolution textures in multiple caching levels (database, main memory and graphics card memory), while making use of coherence between successive frames and combined with frustum culling to generate a visually relevant terrain 3D scene using hardware with limited resources.

Abstract: Welcome to the 11th Communications and Networking Simulation (CNS) Symposium! Ever since the change of name (starting with 9th CNS), this annual event, as part of SpringSim, attracts an increasing variety of participants and papers. The latest trends in Communications Networking are discussed in this symposium. Participants from many countries from all over the world get the unique opportunity of touching bases with each other on what's up in the respective areas of research. It's all happening here. Among many hallmarks of this year's symposium is the ns3 tutorial by one of the developers of ns3. This is probably the first time a tutorial on the latest version of ns3 is presented. The presenter, Professor George Riley of Georgia Tech, is the former Chair of the symposium and knows very well the expectations of the symposium participants. Besides the two-part tutorial, 30 papers have been selected through a peer-review process out of 47 submissions. The quality of most papers submitted was presentable; the top 30 were chosen for a final submission though. Accepted papers happen to be from all sectors; industry, academia and government. Historically CNS (formerly, Applied Telecommunication Symposium) was primarily for industrial research. Soon it became clear that much of the industrial research happens in academia these days. Therefore, by merely looking at the paper content, it is hard to tell whether the research has been carried out in industry, a university or a government laboratory.

Abstract: Contributing to the recent interest in the dynamics of activity-travel patterns, this paper discusses a framework for the dynamic formation of location choice-sets. It is based on the concepts of aspiration, activation and expected utility. Based on principles of Bayesian learning, reinforcement learning and social comparison theories, the framework specifies functions for experience-based learning, extended and integrated with social learning.

Abstract: Agent-based simulation is considered a promising approach for supply chain (SC) planning, configuration and design. Although there have been many important advances on how to specify, design, and implement agent-based simulation, the concerned literature does not properly addresses the analysis phase. In this early phase, SC stakeholders decide what kind of simulation experiments should be performed and their requirements, which considerably influence the whole development process and the resulting simulation environment. This work proposes an agent-based simulation framework for modeling SC systems in the analysis phase. In addition, it proposes a formal method for converting the analysis model into specification and design models. The proposed framework is being validated by means of an agent-based simulation platform developed in the context of the lumber industry.

Abstract: The performance of a typical chemical transport model is determined on two multicore processors: the heterogeneous Cell Broadband Engine and the homogeneous Intel Quad-Core Xeon shared-memory multiprocessor. Two problem decomposition techniques are discussed: dimension splitting for promoting parallelization in chemical transport models, and time splitting, for reducing truncation error. Additionally, a scalable method for accessing random rows or columns of a matrix of arbitrary size from the accelerator units of the Cell Broadband Engine is presented. This scalable access method increases chemical transport model efficiency by an average of 30% and significantly improves the scalability of dimension-splitting techniques on the Cell Broadband Engine. Experiments show that chemical transport models are 31% more efficient on the Cell Broadband Engine when only six accelerator units are used than on a shared-memory multiprocessor with eight executing cores. Our fully-optimized models achieve an average 118% speedup on the Cell Broadband Engine, and an average 87.5% speedup on a shared-memory multiprocessor with OpenMP.

Abstract: Various advanced disease-surveillance models have been developed to provide early detection of infectious disease outbreaks and bioterrorist attacks. New methods that increase the overall detection capabilities of these systems can have a broad practical impact. This paper considers the problem of learning the Contagion Parameter (CP) in a black box model involving healthy, sick and contagious individuals. We base our study on a well-established model of contagion that is characterized by certain fixed parameters, some of which are known, while others are assumed unknown. In the modelling process, we assume that the individuals randomly move within a discretized grid, possibly infecting people or getting infected if they come in contact with healthy/sick individuals. In our study, the parameter of interest involves η which is the probability with which an infected person will transmit the disease to a healthy person. By invoking a novel learning strategy, we show how the CP can be computed using a Training and Testing phase. The results obtained by simulations are very impressive, and are pioneering to the best of our knowledge. The policy-related implications for the contagion control and disease outbreak are also open and very challenging.

Abstract: This paper presents a case study of Web-based distributed simulation across the Atlantic Ocean between Canada and France. The distributed simulation engine, known as DCD++, extends the CD++ environment to expose the simulation functionalities as machine-consumable services based on the DEVS and Cell-DEVS formalisms and commonly-used Web Service technologies. DCD++ provides a platform that represents a step further towards transparent sharing of computing power, data, models, and experiments in heterogeneous environment on a global scale. Also, the simulation service can be easily integrated with other services such as visualization, network management, and geographic information services in a larger system. Experiments have been carried out to investigate simulation performance over commodity Internet connections, and major bottlenecks in the system have been identified. Based on the experimental results, we put forward several areas that warrant further research.

Abstract: The growing popularity of Networks of Workstations (NOW) in scientific computation has drawn increasing interest from the M&S community. This paper addresses the issue of parallel discrete-event simulation of DEVS and Cell-DEVS models on a Microsoft Windows-based cluster system comprising interconnected general-purpose personal computers. We present the architecture and features of PCD++Win, a parallel simulator that takes advantage of the multi-purpose graphical user interface of the DeinoMPI middleware for construction of ad-hoc PC clusters and configuration of simulation environment. This environment significantly reduces the learning curve for general users and the cost of the simulation platform. PCD++Win has been developed using a modular approach that promotes code reuse and allows for easy switching to other middleware technologies. The portability of the simulator is enhanced with multi-platform programming and compilation techniques. Moreover, it leaves open the possibility of further extensions such as Web-based distributed simulation and database-based model construction by leveraging the native support of Microsoft Visual Studio. The experiments demonstrate the capability of the new simulator, making it an ideal M&S toolkit for tapping the computational power of general-purpose desktop computers.

Abstract: Proactive contact campaigns play a growing role in modern contact centers Their traditional usage as a telemarketing tool now is widely extended to different types of service notifications In this paper we describe the most common model of a proactive telephone contact campaign, suggest several methods of its pacing algorithm and present simulation results