PostgreSQL Reviews

PostgreSQL provides the perfect combination of freedom, reliability, stability, security, performance, power, and ease of use. Freedom because it is the only world-class RDBMS not owned by a company. Reliability because anyone looking among the tons of public bugs from other RDBMSs such as Oracle or DB2, and does a comparison with PostgreSQL bugs, will feel confident. PostgreSQL is secured by design, with their own firewall for host-based authentication. PostgreSQL has a superior optimizer, the performance is very good in heavily loaded environments, and it has unique features such as partial indexes and full text search support with GIN indexes. Regarding power, PostgreSQL is a best of breed RDBMS in terms of SQL capability, procedural languages (many of them), programming language and datatypes support (including JSON) and async LISTEN/NOTIFY. Finally, the ease of use or autonomy, is a very valuable factor; PostgreSQL have all the configuration in two single files (postgresql.conf and the "firewall" pg_hba.conf).

We have migrated some databases from Oracle to PostgreSQL lowering licensing fees and reducing the vendor lock-in, gaining sovereignty.

The are many items in my wish list - parallel access (in development), incremental backups, standard MERGE statement (there is a similar INSERT/ON CONFLICT DO), snapshot queries (SELECT AS OF TIMESTAMP), better monitoring stats views, a "trace" feature for a specified running backend, allow streaming replication for a standby from a previous major release in order to minimize downtime during upgrades (this could be difficult with physical standbys, maybe could be easier with logical replication). Of course, everybody needs to keep in mind that any wish list needs to point to the core features, because PostgreSQL is FOSS, and there are other ecosystems out there developing third party features. Here, my wish list for third party software factories is the HTTP->PL/pgSQL adapter, that will enable the development of an Enterprise Web Application Development Framework with just a database (like Oracle APEX).

I've used this solution for more than six years. I started using it in 2009.

Deployment is very easy in PostgreSQL.

Regarding stability, I never viewed runaway processes, nor needed to execute kill or use the Linux IPCRM command against PostgreSQL resources, unlike Oracle and DB2.

It is possible to scale-up (simply adding resources), on demand and without restrictions because PostgreSQL is free. Further, it is possible to realize read scalability with hot standbys.

Very good, based on the few opportunities where I have had to post a problem. I always got an answer quicker in the PostgreSQL forum than opening Oracle SRs or IBM APARs, and the resolutions were also more accurate.

I have more than 10 years of DBA experience, working with Oracle, DB2 and MySQL too. All of them have pluses and minuses, and I usually need to explain that nobody can claim "X is better to Y", at least with the universe limited to world-class RDBMS like PostgreSQL, Oracle, DB2 and MySQL. Of course, certain scenarios can mean that one RDBMS is better positioned than another to handle.

The learning curve is much friendlier with PostgreSQL than with Oracle or DB2. In fact, this could be a risk, because heavily loaded production environments usually needs DBA monitoring and intervention.

I did it in-house. The only advice is to read and learn from PostgreSQL documentation and related books or blogs. PostgreSQL is easy to start, but is a complex RDBMS with many screws to adjust.

There is no pricing/licensing at all. Of course, there are many companies offering PostgreSQL support at different costs and some specialized in several languages/countries.

Ask the community. I believe any committed PostgreSQL user will be happy to help and provide advice.

We use it to convert the databases. We also use it to make a request from my PMS with the Oracle-based server through the third-party interface.

We use this solution for the interface between our property management system and the cloud-based system. It works from both sides.

It is very simple to manage.

It is easy to use.

My main concern is that I can use it to connect with the interface. It is simple, it's good.

It needs more integration.

I have been using this solution for two years.

I have not explored the scalability of this solution yet.

I will continue to use this solution but I don't have any plans to increase our usage.

I have not contacted technical support.

The initial setup is straightforward. It was easy to install.

It took only one day to install, and two to three days to connect with the database.

We have a team of three, two admins, and one engineer to maintain the solution.

We had help from a company from Norway, one from South Africa, and one other from Dubai.

We do not pay for licensing.

Depending on the type of work the user wants to do with the interface I would recommend this solution or not.

I would rate PostgreSQL a five out of ten.

PostgreSQL does a great job with scalability and performance. Since we are using Django/Python – it works great with PostgreSQL as well. The current features that really help us are Full-Text Search, Array Fields, and JSON Fields.

By using PostgreSQL, it was easy to get all the developers configured and running. With using it as our back-end for data storage, our company isn’t worried about data loss or data integrity, especially with snapshots and WAL. Our company is also able to run reports and do analytics off data in custom tables.

The only thing I would really want/need to improve is pgAdmin3. pgAdmin3 is a nice tool for SQL Admin and Queries. But, I wish it would allow multiple queries to be run and displayed in the output pane.

Also, it would be amazing if Full-Text Search would work on encrypted data, but I don’t know how that would be possible with any kind of speed.

In regard to pgAdmin 3, there are so many times I’m in one window running different queries. In Microsoft SQL Server, you can select and execute multiple queries and they are shown in the output pane as separate sections. In pgAdmin3, I have to open multiple windows and execute separate queries and jump from window to window looking at / comparing results.

For the Search/Encryption, PostgreSQL supports having certain fields encrypted in the database (https://www.postgresql.org/docs/9.5/static/encryption-options.html). It also supports Full Text Search (https://www.postgresql.org/docs/9.5/static/datatype-textsearch.html). But, if we encrypt the fields where the lexemes are stored, search won’t work. (Some of the lexemes could give too much information if a user got their hands on the table.)

I have used this solution for three years.

So far, we haven’t had issues with deployment, stability or scalability.

Everything we have needed, we’ve been able to find on https://www.postgresql.org/ or StackOverflow.

We had used Microsoft SQL Server in our previous company. But, the licensing terms kept getting more and more onerous. We started looking at PostgreSQL from the reviews and recommendations, especially since it was open source and Amazon RDS started supporting it.

Initial setup was very straightforward. EnterpriseDB has a great graphical installer to help install on our Dev machines (MacBook Pros). And, AWS RDS makes it is very easy to get set up as well. It allows us to spend our time developing, where our expertise is.

Our initial implementation was done in-house. As long as you follow the step-by-step directions, everything worked great. We also have a managed services company that helps manage the PROD/Staging configuration as well.

Pricing/licensing is the reason we went with PostgreSQL. As a startup, every dollar counts and PostgreSQL has enabled to us to have an enterprise solution at a fraction of the cost of Oracle and Microsoft SQL Server.

PostgreSQL is an amazing product. Very easy to implement and there are many libraries to make it easy to use PostgreSQL as a back-end database solution, no matter what coding language you are using.

This database is highly reliable and offers excellent disaster recovery. In short, I can depend on it to be available and operational.

As we’ve used this since our founding, I cannot speak to how it has improved any function.

The development team has been teasing me with talk of multi-master capability for some time. Their latest release, 9.5.3, does offer some asynchronous capability, but I really want synchronous multi-mastering as that would allow horizontal scaling with much more capability.

Current bidirectional support is limited to the functions provided by a third party integrator. The extension does not offer real-time update guarantees nor does it support DDL updates (schema changes). This is quite unlike the current unidirectional replication scheme, where DDL updates are processed, and there is a way to guarantee real-time updates so the slave server can function as a hot standby. With BDR (bidirectional replication) this is not possible. I understand that in all respects, true bidirectional replication is the holy grail of databases as would give any server in a cluster the ability to take over as a single master without any transaction loss.

I've used this solution for more than 10 years.

There were no issues with the deployment.

I have in the past before the code base had stabilized. Certain procedures just didn’t operate reliably and some internal parameters were just too small and caused problems with large data tables.

It's been able to scale for our needs.

Well, it’s open source so mileage varies, but generally I haven’t had problems. Of course being open source, I can and have looked through the database source code and found the answers to problems I was experiencing.

I have used MySQL, another open source product now controlled by Oracle. It’s reasonably fast and OK for simple databases, but it lacks transactional isolation and its replication setup does not enforce a true master/slave configuration.

It is rather complex to set up correctly. The configuration file is over 400 lines in length and many parameters have only vaguely defined suggestions. Changing a single parameter can have unintended consequences.

Strictly in-house, and I would say anyone attempting implementation should hire the expertise to get the system correct.

Obviously pricing and licensing is non-existent, but the costs need to include the expense of in-house expertise, either employees or consultants.

PostgreSQL is an enterprise capable database very similar to many commercial offerings, but be prepared to either pay for consulting or a long period or experimentation to get the configuration proper.

• Stability
• Standard SQL for ease of development with pre-existing knowledge

Postgres is for our systems a very solid database, easy to use in development and easy also to manage in production environment, and that's valuable because in the end it cuts a lot of costs (also during the upgrading processes).

The Postgres team is already doing an amazing job in putting in features to ease the life of developers and sys admins, such as JSON support and database replication.

A feature that is still a work in progress and it would be nice to have would be native master-master replication, we don't have a need for it right now but it's something that should be addressed in the future.

I've used it for five years.

There have been no issues with the deployment.

We have had no stability issues.

It has been able to scale it for our needs.

I've had no need to use any.

I've had no need to use any.

I used MySQL and Oracle previously, PostgreSQL is open source and fully ACID and SQL standard

It was easy.

We deployed it in-house.

It's a great database engine that is easy to use, powerful, and scalable.

PostgreSQL is the only open source and multi-functional database solution out in the market today. I have used it to replace Oracle (RDBMS) up to MongoDB (Document DB). Postgres has very rich SQL and NoSQL features that gives it a special place in database industry. It’s easy configure and manage features helped to increase adoption in last decade or so.

PostgreSQL is the most flexible database I have worked in my career. For example, when I need geographical data processing for my application, I can use PostGIS. It is the best GIS solution out in the market, by installing extension the extension onto the database. Additionally, PostgreSQL's full text search functionality helped a quite a number of times to avoid introducing another system for text search and use PostgreSQL.

PostgreSQL has introduced number of new features to achieve out of the box multi-master replication and reach JSONB features, horizontal and vertical scalability and parallelization features. I’d like to see them to be production ready. As the number of users of application grows, the multi-master replication will helpful to replicate systems across the continents. We have started using JSONB in number of application, I would like to see some extended features to make JSONB data type more reachable. The horizontal/vertical scalability & parallelization features will help our big data analytics processes faster to improve application response time.

I've used it for approximately 10 years.

I have experience working with over dozen of database and datastores in production environments but PostgreSQL has been the most stable database I have worked so far.

It is missing some key scalability features such as sharding and automatic failover, but the Postgres developers are working on adding these features.

Postgres community is very thriving and Postgres ecosystem is very strong. The product is open source but it has the best documentation and readable code among other open source products. The turnaround time for logged bug is much faster than some of the commercial vendor’s support team. Additionally, there are a number of PostgreSQL professional support companies in the US and all over world.

I used to be an Oracle DBA over decade ago. I have used over dozen of database solutions ranging from MySQL to MongoDB in the distributed production environments but Postgres provides superior features but not limited to; extendibility, performance, scalability, analytics queries.

PostgreSQL is easy to setup compare to any other commercial RDBMS out there in market. It is easy to install and configure.

Postgres has great documentation and various open source tools available for maintenance. However, it is recommended to work with Postgres consulting services companies to expedite design and implementation.

PostgreSQL is truly open source database system so the licensing cost is zero. If you are migrating from any commercial database system, you are saving licensing and recurring support costs.

If you don’t have experience with PostgreSQL, I recommend to attend some of the local PostgreSQL user group meetups and/or conferences to learn how it is being used. If you need help discussing about the solution, you can reach to any of the Postgres consulting services company.

• Security
• HA and DR
• Performance
• Maintenance

I worked as a consultant/trainer. I have a few installations of Dev/Test in my VMs, and I don't have anything in the production installation in our organization.

It's not very user friendly, as it is open source and has no immediate support. You can download the Enterprise version of Postgres that is a GUI installation and also have a support contract.

I've been using it for five to six years. I did the PostgreSQL DBA’s training and led the implementation and data migration projects.

We also use the EDB Postgres Plus Advance Server v9.3 (EnterpriseDB). We use it to evaluate the Database migration from Oracle to Postgres. The data level migration was successful but for objects, and at the security level, you have to fix issues manually.

We've had no issues but setting up the replication, security, and login is different from other databases.

We've had no issues with the stability.

There have been no issues with scaling it for our needs.

As mentioned earlier, there's no support contract for open source but with EDB (EnterpriseDB) you have the open to purchase the support contract. I don’t have any experience with EnterpriseDB support.

It's straightforward, as it's an open source product.

We did it in-house.

As it's open source the ROI is quick.

The core product is free but you have to pay for third-party utilities / plugins and support.

I have evaluated MySQL and Ingres previously and then PostgreSQL. My personal choice is PostgreSQL. I also worked on a migration project from Ingres to Oracle.

• Positive - 100% free as it's open source
• Negative - it's hard to find expertise

If you are planning to migrate an existing environment, please go through with the full UAT cycle at data, connectivity, objects and apps level.

The most valuable features are a good feature set, understandable, and good tuning features. It is, also in our experience, the best open source GIS enabled database, with good support from other open-source tools.

The product has easy installation procedures for the OSes we are working with and gives also a good possibility to tune the database for the given usage.

The database is used to set-up customers projects and is our standard database which we use.

One of the features it needs is a tighter integration with business intelligence software, so that the optimisation of the requests can be handled natively in the database. These features are on the roadmap, and are already partially released in the latest version.

We have been using this database for about 18 years.

We have had no issues with the implementation. Most of the issues we have encountered in external installation where due to misconfiguration.

There have been no issues with the stability. Most of the issues we have encountered in external installation where due to misconfiguration.

We had no issues scaling it for our needs. Most of the issues we have encountered in external installation where due to misconfiguration.

We have up until now not needed additional customer or technical support, these can be purchased by supporting companies.

We have been using Oracle software, and in some cases MS SQL server. The price difference between the used products and PostgreSQL was one of the issues. The second was the ease of installation and the availability on different platforms.

The initial setup is easy to perform, but the tuning is a bit more complex. The installation support on Linux flavours is excellent and has native installation packages.

Implementation and installation was carried out in house.

The pricing of the database is open-source and our investment in the product is based on training.