Development of a Framework for Optimal Utilization of Computational Resources in Distributed Environment

Abstract

Satellite images are large in size and have complex datatype.Serial processing of satel- lite data is very time consuming.A single machine does not have sufficient computa- tional resources to support large scale satellite image processing.In this project,idea is to alleviate this problem using distributed processing framework.This framework helps in parallel processing of satellite image data in distributed environment. Mul- tiple datasets are processed simultaneously depending upon the cluster size under application software through this framework. In this report,I present the design and implementation of distributed processing framework. User can easily pass their appli- cations without explicit use of underlying data distribution mechanisms. Shared and distributed memory is exploited to achieve high performance on each node, thereby minimizing overall execution time. Message Passing Interface Standard (MPI Stan- dard) and POSIX threads are the heart of this parallel framework. Experimental results are presented.