Enhancing Job Scheduling of an Atmospheric Intensive Data Application

Nowadays, e-Science applications involve great deal of data to have more accurate analysis. One of its application domains is the Radio Occultation which manages satellite data. Grid Processing Management is a physical infrastructure geographically distributed based on Grid Computing, that is implemented for the overall processing Radio Occultation analysis. After a brief description of algorithms adopted to characterize atmospheric profiles, the paper presents an improvement of job scheduling in order to decrease processing time and optimize resource utilization. Extension of grid computing capacity is implemented by virtual machines in existing physical Grid in order to satisfy temporary job requests. Also scheduling plays an important role in the infrastructure that is handled by a couple of schedulers which are developed to manage data automatically