Evaluating the performance of SQL and NoSQL databases in an IoT environment

Evaluating the performance of SQL and NoSQL databases in an IoT environment

연구 분야: Databases

학회: 2020 3rd International Conference on Emerging Trends in Electrical, Electronic and Communications Engineering (ELECOM)

Databases play an important role in Information and Communication Technologies and data has no significance unless it is converted into meaningful information. Relational or SQL disk databases where data is structured in tables have served many applications over the years. The increasing need for faster data access in real-time systems has changed the traditional Disk-based database into an In-memory database where data is manipulated and stored directly in the main memory thereby reducing disk input/output operations. With the emergence of IoT and Big Data, NoSQL databases now represent an alternative to traditional relational databases in terms of simpler design and faster operations on large data volumes. This work compares three types of databases namely Relational On-disk, Relational In-memory and NoSQL On-disk/In-memory in terms of Insert, Select, Update and Delete operations together with Indexes, Stored procedures and Triggers in an IoT environment. A plant environment has been monitored in real-time using sensor data such as Temperature/Humidy, Moisture and Water level. Data was stored in different databases and the time taken to execute various operations were recorded. Experiments using the water level parameter demonstrate there is no specific database that provides the best performance for all data operations.