Grid computing is being adopted in a major way by web developers to make way for the next generation of web services. It is being called as the third wave of information technology development which will make a major part of tomorrows connected networks.
In the 1990’s a group of German engineers connected about 100 PC’s together with a nascent version of Linux. At that time these engineers knowing very little could tell that what they have achieved was going to change the way future of the computing world forever. These data grids have been developed so widely that one all data centers have thousands of computers connected in a grid and working on a particular task. The huge processing power provided by the Massive Parallel Processing power of the connected grids is sufficient to handle complex tasks that usually take days to do. To cool these massive cores running is parallel, datacenters had to utilize cooling systems. Server Virtualization was made possible by these data grids but which did not pick up speed because of increasing complexities of Massive Parallel Processing for which the software developed was not sufficient. Companies began spending huge amounts on data grids which promised to improve the performance of the company. The data grid technology gave the power to data centers to distribute processing task to smaller computers with the added advantage of scalability and flexibility. The who processes of data grids is based on two processes, the share everything and the share nothing policy. These two policies have been used by different companies to base their data grids on. The share everything based computing is dominated by the companies SyBase and Oracle. The share nothing category is utilized by Microsoft in its SQL Server Federation Data Store. The cost saving and the increase in power has been the accelerating factor in the area of data grids and this is what will be the driving factor in the future.