Optimize task distribution in grid computing
Abstract
Grid computing is designed to use free cycles of computer to perform large calculations by using free cycles of computer. Grid computing does not need dedicated computers to perform calculations instead it uses free cycles of computer in a network. It works like a virtual super computer, but it doesn’t involve extra hardware cost. As, it’s a cost effective its popularity is increasing day by day. Now a day’s popular applications are being made to support grid computing. For example Oracle database 10g is designed to support grid computing.
Although grid computing is so popular, but implementing software for grid based system is tough. It needs task distribution among computers. So it uses different programming techniques and needs to use special API. The software we are currently using may not support grid environment. Software companies need to convert their programs to support grid environment that involves development cost. On other hand users are not able to use their current software in grid environment. So, they need to buy another program that involves extra cost.
The objective of this project is to run traditional programs in grid system without any modification, so that we can run any executable program in grid system with parallel speeding performance. It will increase program compatibility and reduce cost.