Abstract: The huge amount of streaming data available nowadays are due to massive use of new technologies. Such tremendous amounts of data add a great deal of challenges to the traditional relational database paradigm. The challenges are: performance in reading data and scalability (that is the ability of a database to handle changing demands by adding/removing resources). However, the use of NoSQL database in new enterprises is not a major issue because the new application design will be based on NoSQL database. But the problem appears when the existing systems that relay on relational database are restructuring their systems to implement NoSQL database. They need to reanalyze the system requirements to build up the new database schema. This study aims at developing a mechanism for mapping relational database schema to NoSQL schema and query rewriting of SQL to NoSQL queries. Conceptual mapping has been used which involved the concepts of a relational structure (Table, Simple Attribute, Primary Key (PK), Foreign Key (FK)), and NoSQL concept (Document Collection, Document, Field, Embedded Field, Field List, ObjectId, DBRef). Relationships were realized by defining object references between objects belonging to documents. Mapping algorithm used the metadata stored in the MySQL system tables. The system was implemented in PHP. Experiments were conducted on MySQL and MongoDB based on migration speed and query speed. The read speed of each database was tested, four queries were conducted to pull the number of rows from the product table. The average of these five queries were taken for each implementation. MongoDB proved the faster database management system with a time of 1.78 seconds as compared to 3.35 seconds for MySQL
|
DOI:
10.17148/IJARCCE.2021.10705
