ABSTRACT
Grid computing has emerged as an important field synonymous to high throughput computing (HTC) [1]. Contrary to other systems where the focus is to achieve greater performance measured in terms of the number of floating point operations the system can perform per minute, the importance of grids is defined in terms of the amount of work they are able to deliver over a period of time. The difference between high performance computing and high throughput computing has been illustrated by the Condor project [2]. Grids cannot be considered as a revolutionary technology. Rather they have evolved from existing technologies such as distributed computing, web services, the Internet, various cryptography technologies providing security features and virtualization technology. As we can see, none of these technologies is completely new. They have existed for quite some time and have been serving various needs. The grid technology takes features from these technologies to develop a system that can provide computational resources for some specific tasks. These tasks can be the simulation of stock markets to predict future trends, scientific research such as prediction of earthquakes or serving business needs for an organization having a geographically distributed presence. So in short, grid is an evolutionary technology, which leverages existing IT, infrastructure to provide high throughput computing.
