You would need to make changes in the source code to do that, for something for which I can see no real advantage. Have you got a set of use cases that only MongoDB can implement?
Off the top of my head, you would be far better off using something like Postgresql which is supported out the box, and then using it's built-in support for partitioning the table across multiple volumes, or investigating using an approved extension such as Citusdata or TimescaleDB for sharding or time-series data support.
If this is to support analytics, then you will need to be able query using both time and geospatial functions. Again, Postgresql with PostGIS has the most complete and performant set of geospatial data capabilities and functions compared to any other open source DB. In our tests, it is far ahead of, for example, even SQL Server with geo extensions.
However note that Traccar does not support PostGIS out the box. We implemented our PostGIS-enabled DB separately as we aggregate data from several systems into it.
Postgresql (and several other SQL databases) has (have) been around a lot longer than MongoDB and supports a far broader set of DB functionality.
Hi, I'm very interested to this, I wanna replicate on real time the last position of each device, after this is saved in MySQL I need to send to MongoDB, because I have an app which use this kind of realtime database. This is the same as if I wanted to use Firebase.
Please let me know your thoughts.
Is it possible to redirect and store only positions data in Mongo DB ?
We can keep relational data as it is and just storing the positions data in Mongo DB.
If so we can easily apply conditions to fetch attributes data using Mongo DB Filter and
also, this engine supports large data sets.