Parse Database

You can use MongoDB and Postgres on your parse server as database. Although MongoDB is a preferred database option with parse server but to have a more stable schema in your new project Postgres is better to go with.

MongoDB

Let’s familiarize with MongoDB first, if you are going to use it for the very first time before proceeding. For Parse Server, MongoDB requirements are given below:

• An SSL connection is better to opt.
• MongoDB 3.6 version. If you are setting up MongoDB instance production first time, then its recommended to go with either ObjectRocket or MLab. These Database as a Service companies are offering managed MongoDB instances which you can easily scale when needed. However, indexes management is completely up to you. To run MongoDB on your own infrastructure, the RocksDB Storage Engine is recommended. However, to allow your data layer scaling direct queries for operations to MongoDB secondary.

Postgres

Requirements of Postgres for Parse Server are:

• PostGIS extensions 2.3
• Postgres 9.5 version When you are going to pass a right Postgres URL then its database adapter will be loaded automatically, just like: postgres://localhost:5432. Keep in mind: Here we have some of the most important cautions which you must keep in mind while starting working with any of your preferred database.
• Join tables will be resolved in memory while no performance improvements are there for pointers or relations by using Postgres over MongoDB.
• Mutating the schema may imply running an alter table, therefore, it is highly recommended to run your schema when your tables are having some space.
• If you want to maximize your performance then it is always better to index your tables in a proper way.