A SEMANTIC BASED SCHEDULING ALGORITHM FOR DATA INTENSIVE APPLICATIONS ON GLOBAL GRIDS

Due to the increasing size of data intensive applications, the datasets are need to be stored in a distributed manner and multiple copies of data sets has to be replicated to provide easier access to the grid applications. In these kinds of systems, scheduling is a key challenge because of the location of data and heterogeneity of a node by holding multiple types of data sets in the same location. We propose a promising technique to schedule the grid applications, using semantic based classification. Our algorithm classifies the grid resources according to the semantic meanings of data sets it holds, using which it identifies the data location which are necessary for the grid applications to be executed successfully