Service oriented distributed manager for grid system

Current problems in science and engineering become more complicated and need more computing power to tackle and analyze. The processing power of a single computer system has become inadequate for that problem and use of a supercomputer is not always an optimal solution. Nowadays institutions as universities or companies have lots of Desktop PCs in their infrastructures. With advanced in hardware, software and network gird computing are developed to benefits from these futures in idle computers in an efficient way to solve complex problems that need powerful computing requirements. Grid technology, which connects a number of personal computers, can achieve the same computing power as a supercomputer does, also with a lower cost. The goal of grid computing is to aggregate idle resources on the Internet or Intranet such as CPU cycles, storage spaces, etc. to facilitate utilization. In this research we present computational grid framework that used to utilize idle computers in Intranet to execute jobs in an efficient way than a single computer. We design and implement Service Oriented Distributed Grid Manager (SODGM) framework based on web services technology to manage resources and jobs in the system. SODGM balance the load among available resources to increase computers utilization in the system and minimize jobs response time. To evaluate the SODGM system, we performed tests on the Computers Laboratory in the Faculty of Computing and Information Technology in King Abdul Aziz University. The tests show that SODGM system provides good performance compared with using single computer in terms of execution time, resource utilization and system throughput.