Support for vectors, XML, JSON, UUID and the functions and operators for them are useful.
- Very good indexing solutions, especially for arrays and text search on large quantities.
- Very good behavior on large databases (>800GB) and large tables.
- Datatypes - support for vectors, XML, JSON, UUID and the functions and operators for them are very useful.
- Programming languages - I deeply appreciate the extensibility of the server to allow more than one language for scripting functions. We mostly use PL/SQL, but recently I've tried PL/v8 and was astonishing to see that it works.
- Execute with Query Parameters is a big plus for us as there are no type conversions needed to run dynamic queries reducing the run time by approximately 25%
- There is a huge amount of community interest in developing and supporting the product.
Improvements to My Organization:
All of our BI needs are implemented using PostgreSQL functions. Our tools are designed in such a way that the BI is implemented as much as possible through our database. We need a lot of flexibility to extend the BI logic on a specific installation.
Room for Improvement:
Synchronous replication is needed, and although I have only searched the web for a couple of hours the only solutions I could find were at the application level of the database.
XML and JSON manipulation functions need to be improved as for some applications we need to add new elements to these "structures". To do this with JSON I currently use a PL/v8 extension.
We have had no issues with the deployment.
There have been no performance issues.
It's been able to scale for our needs.
Try it out! Sometimes, due to the high degree of configurability and extensibility, you might see some issues arise, but it's not that difficult to figure out what's happening. You will definitely be amazed by the SQL features and extensibility.