Evaluation of NoSQL Databases Features and Capabilities for Smart City Data Lake Management

Abstract

A data lake means there’s an immense data resource or repository. Data lake stores enormous data and uses advanced analytics to pair data from various sources with different types of structured, semi-structured and un-structured information. The lifeblood of a smart city is data. Effective data management is not limited to data collection and storage, but must also involve shared and combined data so that it can be accessed, analyzed and used across agencies, within organizations, and even across the society. NoSQL is a form of database that is becoming increasingly common among web firms. NoSQL databases are non-tabular and store data rather than relational tables in a different way. NoSQL databases come in a variety of forms, mainly document, key-value, wide-column, and graph based on their data model. NoSQL offers easier scalability, better performance compared to conventional relational databases, and consists of many data types, such as document, key-value, wide-column, and graph. This work studies NoSQL database features and capabilities by considering four indicators, namely performance, scalability, accuracy and complexity, in order to measure the compatibility of NoSQL databases with multiple data types. The result of the experiment reveals that when accommodating massive data volumes, MongoDB is the most stable NoSQL database.