Evaluation of Cloud Databases as a Service for Industrial IoT Data

Abstract

AbstractApplications using IoT sensory data, such as in Industry 4.0, are a classic example of an organized database. This paper focuses on evaluating three types of DBMS, MongoDB, PostgreSQL using JSON and the relational PostgreSQL, measuring average, jitter, and loss of response Time and achieved throughput. Three scenarios were thoroughly tested, (i) data insertions, (ii) select/find queries, and (iii) queries related to correlation functions. Experimentations concluded that MongoDB is between 19–30% faster than Postgres in the insert queries, achieving 51–55% higher throughput. Additionally, relational Postgres is x4 times faster than MongoDB and x2 times faster than Postgres JSON in the selection queries, achieving 31–35% higher throughput. Finally, the two versions of Postgres performed equally concerning response time in the correlation function queries, while both of them outperformed MongoDB by x3.6 times. Contrariwise, in the correlation function queries, MongoDB achieved 19–24% higher throughput than both versions of Postgres.KeywordsDatabase systems performance evaluationMongoDBPostgreSQLIndustrial systemsIoT dataIndustrial IoT