Effect of Grid Job Slice Variation on Virtualized Desktop Grid Performance

Grid computing has come a long way in solving data and compute intensive problems. There have been significant gains in solving these problems using grid based systems by dividing these jobs into sub jobs and submitting sub jobs on various nodes in the grid system. Various job characteristics may have effect on the execution of jobs on grid systems. The performance of grid systems varies with the type of problem taken. In this paper, the grid systems are studied for variations in the size of job fragments of sub jobs. With different size of sub jobs the grid systems responds differently. As the job fragment size is increased the performance of grid system is observed to increase. The job taken is video conversion from AVI to FLV.