ABSTRACT

The grid is rapidly emerging as the dominant paradigm for wide-area distributed computing [202]. Its goal is to provide an environment for coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organizations [211]. Most of the current grid deployments have focused on data intensive applications where significant processing was done on very large amounts of data [198]. The data required by such applications is largely distributed in various storage systems. The need to access remote data with “near-local” performance is crucial for scheduling and managing of application execution.