Design and Development of Grid Portal

Abstract

Grids are becoming platforms for high performance and distributed computing. Grid benefit users by permitting them to access hetrogeneoun resources such as machines, data, people and devices that are distributed geographically and organizationally. It benefits organizations by permitting them to offer unused resources on existing hardware and software. They allow users to execute compute intensive problems whose computational requirements cannot be satisfied by a single machine. Grid Computing has emerged as a new and important field and can be visualized as an enhanced form of Distributed Computing. With the advent of new technology, it has been realized that paralleling sequential applications could yield faster results and sometimes at a lower cost. Possessing multiprocessor systems was not possible for everyone. Thus, organizations started looking for a better and feasible alternative. It was found that though every organization possessed a large number of computers, which meant that they had huge processing power, but it remained underutilized. A new type of computing then came into existence and was known as Distributed Computing. In Distributed Computing, the problem to be solved is divided into numerous tasks that are then distributed to various computers for processing. The computers being used are connected through a Local Area Network (LAN). Also the problem needs to be divided into modules that can execute in parallel to each other. Parallelism can be either Data Parallelism or Functional Parallelism. Data Parallelism means that each computer performs the same function but for a different data set; whereas in Functional Parallelism, each computer performs a different function for different or same data set. As more and more resource intensive applications (compute-intensive and data-intensive) were developed, a need for larger amount of resource sharing was felt. Then the Grid computing comes into existence. A Grid enables the selection and sharing of a wide variety of geographically distributed resources including supercomputers, storage systems, data sources, High Performance Computers (HPC), and other resources owned by different organizations for solving compute-intensive and data-intensive problems.