Weather forecasting using parallel and distributed analytics approaches on big data clouds

Abstract

In cloud environment, big data analytics is a very innovative idea; in this paper we have design architecture for parallel and distributed analysis of big data in cloud environment. We can use this architecture for every field in which we have a problem with data analytics. We are going to use this for meteorological data because meteorological department have a huge data which is related to weather. Weather forecasting refers to predicting future weather conditions on the basis of available data. These weather conditions can be utilized to prepare ourselves for future as well as alerts about disaster therefore saves the valuable human life. Keeping these things in mind we design system architecture for weather forecasting. First we use Hadoop to extract big data from NCDC through HDFS. Secondly we process this data through Map Reduce. Finally we get output which includes max temperature, minimum temperature, humidity, rainfall on any future date using past few years data. Analyzing such huge volume of data i.e big data and predicting future temperature brings immense importance to our work.

Keywords:

• Big data
• My SQL
• HDFS
• Hadoop
• MapReduce
• Cloud