As we've seen, there are plenty of options to store our microservice state. Depending on the type of information we're manipulating, there are databases better-prepared to handle our data.

It all depends on a few different questions we should ask ourselves:

• Is our data integrity important?
• Is our data structure complex?
• How and what type of information do we need to acquire?

If our data integrity is important or the data structure is complex, do not use in-memory databases. Depending on the complexity, see if you need a non-relational database, or whether you can go with a relational database that can handle more complex manipulation and data aggregation operations, which will help you to achieve the last point.