Arris Ray

Date of Award





Computer Science - Applied Computing Track


TSYS School of Computer Science

First Advisor

Stan Kurkovsky

Second Advisor

Dr. Bhagyavati


The primary aim of modern computer networks is to permit simplified sharing of resources (e.g., files and data) or to provide access to remote resources (e.g., printers or remote file-systems). Grid networks aim to provide a more sophisticated framework for securely sharing, selecting and/or aggregating a wide-variety of remote resources, whether logical (e.g., files and data) or physical (e.g., remote sensors, processing power or disk storage space). Provided that circuit design continues to improve at the relatively steady pace that it has for the past decade, in addition to standard desktop systems, miniature and mobile devices will soon become substantially significant nodes in extant networks and emerging paradigms of computing, such as grids. Although modern mobile devices, such as tablets and personal digital assistants, are powerful computers in their own right, they are simply inadequate systems for most computer applications, which tend to target the de facto standard of desktop-class computer systems. Mobile devices, by merit of their reduced size and portable nature, are inherently subject to a number of constraints, including the fact that they possess substantially fewer computational resources (e.g. processing power, disk-storage space) than desktop-class systems and that they operate on an independent and limited power supply. Additionally, mobile devices tend to adopt a wireless, rather than wired, means of communication for networking, making them subject to the set of constraints that characterize wireless communication. Wireless communication technologies are generally subject to concerns such as signal interference and unpredictable levels of connection quality or connectivity. These unique constraints have already required re-evaluation and re-design in various aspects of the Internet, from the network level (e.g. Mobile IP) to the application level (e.g. compact, device-specific Web browsers). Because grid networks are anticipated to build on the existing framework of the Internet and Internet technologies, it seems safe to assume that mobile devices will have a similarly significant impact on grid networks. Unfortunately, the majority of current grid applications target clusters of resource-rich desktop systems for the express purpose of scientific or engineering research and there seems to be relatively little information available on the impact of mobile devices on wireless grid environments. Thus, the goal of this work is to develop a simple simulation system to investigate the effects of introducing a population of mobile devices into a grid network. This was done to determine the nature and extent of any effects that mobility may have upon the resource-sharing capabilities of a wireless grid network.