OBJECT ORIENTED IMPLEMENTATIONS FOR DEVS FORMALISM

ABSTRACT

Discrete Event System Specification (DEVS) formalism is a Modeling and Simulation (M&S) framework that provides a means of specifying an object called a system. This work presents implementations of this formalism using the Parallel DEVS simulation algorithm and developed as a new simulation engine (simulator). The presented techniques use an approach that exploits the Object Oriented Programming (OOP) paradigm concepts through Java programming language. Our goals include the implementations of this algorithm, benchmarking and analysis as well as using OOP approach to implement the CDEVS algorithm. Also, the implementations we propose are: one for sequential computers and the other for parallel computers. This is for comparison reasons regarding the real gain in performance (due to the exploitation of parallelism).

CHAPTER 1

INTRODUCTION

1.1 BACKGROUND

Besides the evolution in humans learning about nature, the twentieth century witnessed the creation of many artificial applications such as traffic control systems, automated factories, computer architectures, or biomedical devices that made it difficult to study especially when the complexity or the required level of detail is high. However, Modeling and Simulation (M&S) provide a well-developed, well-proven approach to solving such problem that advances steadily as more computing power becomes available at less cost. There is a need to exploit the computing power of nowadays technologies by distributing simulation on multiple processors to be able to model and predict the behavior of complex systems (Road Transport Network, Weather forecast, Fire spread and many others). This is to reduce execution time, perform real time execution, and integrate simulators. The development of computers has offered alternative methods; models can be executed using computer simulation, allowing users to experiment different conditions under risk-free environments [1]. The advantages with this approach include model reuse, user interactions with the simulator, verification and validation of models and economic benefits. The M&S process begins with a problem that needs to be solved or understood. According to Zeigler [2], the source system is the environment under analysis while the model is an actual representation of such system. A simulator executes the model's instructions thus generating the model's behavior. Discrete-event modeling is based on the notion of event [3], which is defined as a change in the state of the model. An event occurs at a given instant (called the event time) and causes the model to activate in order to produce a state change (e.g., at least one attribute in the model will change). Finally, a model’s state is the set of values of all the attributes of the model at a given instant. The model’s attributes are usually stored in variables; state variables are those that will influence the evolution of the model’s behavior.

Department: Computer Science (M.Sc Thesis)

Format: MS Word

Chapters: 1 - 5, Preliminary Pages, Abstract, References, Appendix.

No. of Pages: 49

Price: 20,000 NGN

In Stock

Buy Now

Our Customers are Happy!!!

See What Some of Our Customers are Saying About Us