Data Migration tool to minimize the spatial redundancy without affecting the query performance

Abstract

Objectives: The study is to minimize the spatial redundancy without affecting the query performance, i.e., time taken to execute the query. Methods/Statistical Analysis: This paper illustrates a new method of analyzing the source database system: data, relations, etc., and generation of a Graph model based upon the database and conversion of the model by creating a new algorithm compatible with NoSQL databases, keeping in mind the following parameters: (1) Less time to execute each query, (2) Maximize spatial performance with minimum space and (3) Robustness of algorithm i.e., no data loss, metadata loss, or relational data loss during two transition stages, that is from old data migration to new model and analysis of new model (comparison with relational model). Findings: The advantage of NoSQL model is the decreased time taken to execute each query and increased speed for execution of queries due to ability to deal with the unstructured data. It was found that out of the three scenarios considered in the study that is nested, non- nested and hybrid, hybrid scenario is the one that possesses the most consistent performance and hence can be stated to be the best performer. Further, the study presented a robust algorithm for the migration of data in such a way that data loss, meta data loss, or relational data loss during transition is null or minimum. Application/Improvements: The present study proposes a method to convert the relational database to NoSQL database schema conversion model. This specific model has used the concept of embedded documents to improve query speed of NoSQL database. Keywords: Data Migration, MongoDB, Relational Database, NoSQL, SQL