Study and Analysis of Various Big Data Analytics Tools for Data Processing

Huge amount of data is generated in this digital world with the size of terabytes and petabytes. The traditional database management system is failed to process these exaggerated data. Big data is the combination of high volume, high velocity and different veracity of data. These data are extracted from multiple sources such as machine data, social media data, medical data, banking data, transaction data. E commerce data and business data. In order to process the huge amount of data, the specially designed software, tools, framework and database file systems are required. Since the structure of big data is considered as unstructured and semi structured, the highly scalable and fault tolerant processing environment is needed to cop up with these big data. The big data analytics life cycle is divided into few steps, such as business understanding, data understanding, data preparation, modelling, evaluation and deployment. The employed software or tools are designed and developed to fit into these life cycle of big data analytics. In this study, the predominantly used Hadoop framework, and mongo DB database are studied and applied on sample dataset to process the larger amount of big data. The results section is producing performance analysis report of these above mentioned data processing tools. Each and every tools are performing well and producing the expected efficient results on those huge amount of data. 2021 tokyo Olympics dataset is downloaded from the kaggle website and processed in this research study. This dataset contains 5 fields of records like details of athletes, coaches, gender entries, medals and names of the teams worldwide.