High Performance Processing of Satellite Data Using Distributed and Parallel Computing Techniques

In today's world of technological revolution when the volume of the data is increasing enormously coincided with the growth in technology, it has become crucial to process and store data adroitly. Due to increasing demand of high processing speed, the traditional methods of processing satellite data have become incompetent. This propelled the need for high performance computing, which is the ability to process data and complex calculations at an accelerated speed effectively and accurately. It takes prolonged time for batch processing of satellite images which acts as the foundation of analysis developments in many technological and geological fields. In this paper, presented, a proposed distributed and parallel computation solutions for satellite image processing and computation of various indices normalized difference vegetation index that improves the performance of the system. By taking advantage of apache spark and cluster computing techniques real-time high-speed stream processing of satellite data is achieved. Some main features are discussed comprehensively about apache spark cluster formation, distributive and parallel computing methodologies, calculation and processing of indices with satellite data of Landsat 5. Also, python programs for processing of satellite data of Landsat 5 are executed and their results are presented in terms of processing speed and time.