I like PostgreSQL because it's a robust alternative to commercial databases. It stands very well against other open source competitors, like the more popular MySQL. It's fast and reliable, managing inserts and updates of tens of thousands of rows flawlesly.
Besides that, it's easy to install and to manage, and it has some handy non-standard SQL commands. Documentation in the official website covers most of the common and advanced tasks with a lot of examples of querys. Review collected by and hosted on G2.com.
Managing everything from the command line can be a little bit cumbersome. There are a lot of third party graphical interfaces to access the databases from the desktop (pgAdmin) or via browser (phpPgAdmin) but you have to install them and learn to use then on your own.
The same with reporting tools, there are a lot of nice ones but you have to search and try what is the best for you. Review collected by and hosted on G2.com.
It's a good and not too heavy solution in terms of relational databases.
It's relatively smooth to install, has several good features and performs well.
It doesn't have too much of a steep learning curve.
There is good online support. Review collected by and hosted on G2.com.
There can be some issues installing it in a Windows environment.
Nothing impossible, but it can take time to solve them.
A total newby to relational databases could find it difficult to use it, especially as there is no GUI. Review collected by and hosted on G2.com.
Highly extensible, several PL implementations, easy to write stored procedures if you already know python, perl or C because it implements them as procedural languages, tons of addons to extend its functionality, functional indexes to improve performance if you need to search by computed values and it makes it really easy to perform queries based on data stored in JSON fields and it supports master-slave replication natively.
It has a lot of tools to check the performance of your queries, the consistency and the health of your database and to collect garbage and clean up your database (EXPLAIN and Vacuum)
If you want a NoSQL data sture, Postgres also have you covered, you can try HStore. Review collected by and hosted on G2.com.
It is too large and has thousands of features, so documentation sometimes is outdated and you can get lost very easily if you don't know exactly what you're looking for Review collected by and hosted on G2.com.
Powerfull and stable db engine, it's cross platform, a great support for business, it has many administrative tools and extensive documentation. Review collected by and hosted on G2.com.
Actually nothing specific, maybe the fact that it's less mature that its contraparts dbm therefor it has less popularity in market carring a little bit resistence to use as principal dbm in actually projects. Review collected by and hosted on G2.com.
I like the NoSQL key-value store (HSTORE) that works out of the box and binary JSON storage. This allows me to keep the SQL-based pipeline I have, yet venture into NoSQL land safely. And all of that for free. Review collected by and hosted on G2.com.
It had a bit of a learning curve to use properly. Review collected by and hosted on G2.com.
I wouldn't consider myself to be a database expert, but from basic usage via Rails apps, I've been very happy with the features I have used. Full text search, custom data types (json, arrays, etc), performance optimizations, extensions (like Postgis for spatial data), sharding and partitioning support... I've mostly interacted with Postgres via ActiveRecord though. Been very happy with it in multiple production instances across apps of all sizes (from tens of users to tens of millions). Review collected by and hosted on G2.com.
When I first started using Postgres it wasn't well known so it was hard to find support for more advanced features... For instance RDS Postgres didn't support replication for a while... That's all been solved now though. I've been very happy with it in the last few years. Review collected by and hosted on G2.com.
Includes all the usual SQL features, and also has support for lots more:
* Full text search: without needing to configure and maintain standalone services like ElasticSearch, PostgreSQL lets you add full text search to your application, with support for dictionaries, stemming, trigram search, and much more: http://www.postgresql.org/docs/8.3/static/textsearch.html
* JSON columns: can act as a non-relational database with JSON support like MongoDB, which lets you do all that the NoSQL databases can, while also having full relational support as always.
* Can work with geographic objects with the PostGIS plugin, to build searchable and fully detailed maps for POIs and such.
* Has hooks for very detailed analytics with the pg_stat_statements extension. Review collected by and hosted on G2.com.
Since it is more strict that other SQL databases, it is hard to make some complex queries which involve joins with groups.
Support for PostgreSQL is pretty high in servers, but operating systems do not support it by default, and so you always need to install it.
I still haven't found a nice GUI to do light support tasks. There's PGCommander for Mac, but it lacks features. Review collected by and hosted on G2.com.
Its simplicity and its ability to work with large datasets Review collected by and hosted on G2.com.
The default configuration is not optimal for the kind of work we do (large datasets and intensive queries). It requires significant tweaking. Review collected by and hosted on G2.com.
The speed in conjunction with the native JSON support. With the on the horizon 9.5 the added features will only deepen my appreciation for Pg. Being able to have all the functionality of a NoSQL type DB on top of a a traditional RDMS makes using something like Couch/Mongo obsolete. Review collected by and hosted on G2.com.
I am not the biggest fan of the vacuuming that is required for larger datasets. Eventually most people hit the dreaded "transaction wraparound" error if they are not careful and it is a nightmare to recover from sometimes. Review collected by and hosted on G2.com.