APPROACH TO A SIMULATION VIRTUAL MACHINE
Need help with a related project topic or New topic? Send Us Your Topic
DOWNLOAD THE COMPLETE PROJECT MATERIAL
APPROACH TO A SIMULATION VIRTUAL MACHINE
Chapter one
1.1 Introduction to Computer Simulation.
A simulation is a calculation that mimics the behaviour of a real or imagined system over time. Computer simulation refers to computations done on a computer.
Simulations are commonly used to analyse system behaviour, such as fire spread, weather forecasting, air traffic control, decoding the human genome, and designing new telecommunication networks, without the need for physical prototypes.
Modelling and predicting complex system activity is challenging due to increased data volumes, statistical requirements, and global user groups.
Distributing simulation across many processors can minimise execution time, enable real-time execution, and combine simulators. For additional information on parallel and distributed simulations, refer to Fujimoto’s book (2000).
“Creating a model is the first step in studying any physical system for simulation. Such a model can be classified into numerous types: There are six types of models: conceptual, declarative, functional, constraint-based, spatial, and multimodel. (Fishwick & Lin 1996).
The conceptual model defines qualitative terminology and class hierarchies within the system. The conceptual model defines the properties, procedures, and characteristics of system components without assigning dynamics to them.
The following four model types focus on system construction, such as Petri nets, Queuing models, and cellular automata. The Multi model type combines fundamental model types to build a model with component models, each representing a level of abstraction for the system (Fishwick and Lin 1996).
2
A simulation model is created by retaining relevant characteristics of a real or imagined system and discarding extraneous ones. This model can then be executed on a computer.
DEVS abbreviating. Discrete Event System Specification is a hierarchical formalism for modelling and analysing general systems, including discrete event systems (state transition tables), continuous state systems (differential equations), and hybrids of both (Wikipedia 2007). The following chapter will cover DEVS in further detail.
1.2 Objective.
Our goal is to create a virtual machine that can simulate a DEVS model on any hardware host, such as a LAN, WAN, Grid, Cluster, or the Internet.
The virtual machine has three layers: the modelling layer receives the DEVS model, the simulation layer simulates the model using either the pessimistic or optimistic synchronisation algorithms, and the Middleware layer maps the simulation to any hardware host.
The virtual machine’s kernel will include both CDEVS and PDEVS simulator implementations, as well as distributed versions.
1.3 Work Done
We implemented a PDEVS for an existing CDEVS simulator utilising meta modelling. The CDEVS and PDEVS simulation packages are available, but the virtual machine’s whole package is not yet available due to its research prototype status.
The CDEVS and PDEVS simulators are Java-based and use JVM as the virtual machine. We tested the simulators using an example and compared simulation durations.
3 1.4 Organisation
Chapter 1 is the introduction chapter. Chapter 2 describes the CDEVS and PDEVS formalisms. Chapter 3 presents the modelling of the simulators using the Unified Modelling Language (UML), including both static and dynamic views.
In Chapter 4, we explain how we transform the Model into Java code, including the packages and classes involved. Chapter 5 provides an example of comparing simulation times between two simulators. Chapter 6 is the final chapter.
Need help with a related project topic or New topic? Send Us Your Topic