NoSQL database

NoSQL stands for "Not only SQL" and refers to database systems that follow a non-relational approach. Instead, they are based on various alternative database models, which makes them horizontally scalable (to large amounts of data) and particularly suitable for Big Data applications. In relational databases, data is stored in tables with columns and rows. NoSQL databases, on the other hand, use value pairs, objects, documents, lists or rows to organise the data.

NoSQL are optimised for applications where relational SQL databases reach their limits. Due to the lack of rigid schemata of relational databases, NoSQL databases can be used very flexibly and are particularly suitable for large amounts of data, as their complete architecture is designed for performance and scalability.

What needs to be considered?

• The data for the NoSQL database should be modelled correctly.
• Some NoSQL databases do not offer security features.
• Clear understanding of the database requirements for the application.
• The chosen database should be able to cover a variety of use cases so that it can be used for future applications.
• If nested transactions are required, which in turn require rollbacks and savepoints, then NoSQL is probably the wrong choice.