Deadline stringency based job scheduling in computational grid environment

Computational grid, which came into existence in mid 1990s, represents a computing infrastructure, which is distributed in nature. Grid, these days, is extensively used in science and technology. As a result, computational grid has become a promising research field. It has become a highly productive platform for dynamic problem solving. Grid provides resource sharing among different organizations. Due to the provision of scalable resources, these days grid´s usage is rapidly increasing in industry also. But, computational grid has few distinct requirements in comparison to those of traditional distributed high performance computing systems. To utilize such grid systems, resource allocation and utilization are key challenges to be dealt with. Henceforth load balancing is another common problem as because the load scenarios of individual resources are dynamic in nature. The objective of this paper is to enhance the efficiency of an already proposed NDFS algorithm to solve the existing problem of load balancing in computational grid. Comparisons of the performances of a few existing load balancing algorithms with proposed NDFS are also presented in this paper.