Item: PostgreSQL
Rating: 9
Author: Verified User

Overall Satisfaction with PostgreSQL

Use Cases and Deployment Scope

In my team, we primarily use PostgreSQL as a relational database for standard use cases, where we for example have some information that we want to keep track of but not having everything in a single table, choosing instead to build several relations between the tables. The other use case we have, which is the primary one, is the basis for our Airflow instances - this is something set up by default by Airflow but for us it is sometimes necessary to extract information from this DB for further processing.

Pros and Cons

Pros

Query performance
Automatic upgrades
Retro compatibility

Cons

Clearer indications on what is the query plan, to optimize the query
More out of the box, Postgres specific, SQL functions
It would be nice to have a more visual aid of the relationship between all tables, but possibly this depend more on the UI used

Return on Investment

0% incidents due to maintenance - the only time was actually caused by an upgrade of two major versions we did
When correctly used, the difference in query performance became 100 times faster (simply due to using indexes and PostgreSQL specific functions)

Usability

There is a learning curve, as it is a very well rounded product which does not hand-hold you: this means that you cannot necessarily get it up, run it and be done with it, however it is not a matter of months but maybe a one week heavy study (for instance, learning how to best optimize your queries).

Alternatives Considered

MySQL

MySQL is an Oracle product which has in itself some known issues due to that (support, contract terms). Based on my knowledge, PostgreSQL support everything that MySQL support (syntax wise) and it adds more improvements and syntaxes that make the life of database engineers and anyone trying to use it much simpler and with a higher performance.

Key Insights

Do you think PostgreSQL delivers good value for the price?

Yes

Are you happy with PostgreSQL's feature set?

Yes

Did PostgreSQL live up to sales and marketing promises?

I wasn't involved with the selection/purchase process

Did implementation of PostgreSQL go as expected?

Yes

Would you buy PostgreSQL again?

Yes

Other Software Used

Apache Airflow, Apache Spark, Amazon DynamoDB

Likelihood to Recommend

Based on my experience, PostgreSQL is exactly what I imagine when thinking of a relational database: it is fast for querying and intuitive (if the underlying relations have been well written) and its writes are also efficient. Of course you need to know that a relational database is what you need, for example I would not recommend it if you have a lot of unstructured data that have no way to create a relation from and if you need to run heavy aggregations over your data, as it is not where PostgreSQL would shine.

Comments

Please log in to join the conversation

PostgresSQL the definitive relational DB