Archived: 11 February 2003
University of Adelaide Honours thesis, October 2002.
Supervisors: Andrew Wendelborn, Paul Coddington and Kevin Maciunas
A grid is a high performance computational environment that is composed of heterogeneous resources spanning wide area networks and multiple administrative domains. This is a new computing paradigm that aims to make high performance computational resources available to anyone accessing the grid.
Problems inherent in such an environment include the ability to discover suitable resources and how to schedule separate tasks of an application on those resources in such a way as to maximize the applications performance. A component of the grid infrastructure of which an instance is associated with each application is called the resource broker. It is the resource brokers task to implement the execution of the application on the grid transparently from the users view.
This project implemented a grid environment simulator incorporating NWS that can be used for the thorough investigation of different resource brokering techniques and the use of dynamic information in that process. An investigation of the different techniques was also undertaken.