How PostgreSQL Became an Overnight Success 30 Years in the Making

451Research recently estimated that around 30 percent of technology companies use PostgreSQL for core applications. This object-relational, open source database system for enterprises is known for its power, reliability and scalability. VMware engineers Peter Geoghegan, Kevin Grittner and Michael Paquier are some of the bright minds behind the growing success of PostgreSQL.

These three members of VMware’s PostgreSQL engineering team have a storied history with the project. Peter contributed sorting improvements, group commits and the UPSERT patch, which allows someone to automatically insert a row or update an existing one. Kevin added materialized views to PostgreSQL and works on contributions around SSI. Michael works on commitfest management, code reviews and bug fixes.

As Peter notes, “the popularity of PostgreSQL is generally considered to be on the up-and-up.” Users hail from a variety of industries and government agencies, which use PostgreSQL when high-volumes of data need to be quickly sorted into custom formats. Currently Peter is continuing his work with corruption detection in PostgreSQL and is researching index bloat and potential solutions. His most recent contribution was a code commit for amcheck.

Peter, Kevin and Michael have been active observers and contributors of PostgreSQL for years. Beginning as an academic project at UC Berkeley in the 1980s, PostgreSQL is now more popular than at any other point in its history. The PostgreSQL experts said a big part of that longevity is due to the tireless collaboration and trust within the open source community.

“This is almost like a lifestyle choice at this point, honestly. That’s really not an exaggeration,” Peter said, talking about his years of experience working on open source projects. If not for its open source contributors, Kevin says PostgreSQL would not be where it is today. “One of the advantages of open source is that if you need something and it’s not there, you can make it happen.”

Make it happen they did—over a lengthy period of time, that is. The prevailing joke is that PostgreSQL is an overnight success 30 years in the making. However, Peter, Kevin and Michael believe success was fueled by a slow-growing awareness leading to a recent surge in popularity.

PostgreSQL

PostgreSQL is based on well-established ideas when it comes to a database system. As an open source project, this familiar foundation is much easier for contributors to navigate. PostgreSQL was also consciously developed to make all enhancements more seamless than traditional methods of development. This enables different people with different needs to utilize PostgreSQL.

For example, a set of Russian scientists needed a new index for their astronomy work with complicated inputs documenting work in multiple dimensions. They were able to write their own code and plug it into PostgreSQL because of the project’s sound framework. They contributed their indexing code to the open source community, and now others are using it for more terrestrial purposes, like running a search for the closest pizza restaurants near someone.

Many of PostgreSQL’s features carved out a path for its success, but picking a favorite is a difficult task for the team. Peter likens this to picking the favorite feature of your car; how can you possibly choose? However, Michael is fond of one of PostgreSQL’s newest offerings. “I like a lot of the work that’s been put into a new feature called SCRAM (Salted Challenge Response Authentication Method), which is the new authentication protocol,” he said. He’s one of the authors of the feature, which has been in development for four years.

Another reason for PostgreSQL’s success lies in the collaborative, dedicated and professional open source community. Peter, Kevin and Michael are quick to praise the community. For Michael, PostgreSQL was his first foray into the open source world. “[It] got me introduced to the basics of the community and the basic way of doing things.” Michael now ranks third among contributors of patches (or fixes) for the program.

Peter and Kevin, meanwhile, are veterans of the open source community. Peter believes the community is driven by a long-term outlook on advancing the project. Kevin is quick to point out the camaraderie in the community and around PostgreSQL. In fact, many of the features added to PostgreSQL came from contributors sharing beers during industry conferences.

The excitement around this project has lured many more collaborators, with approximately 200 developers currently working on PostgreSQL. The recent growth in adoption of the database is inspiring even more developers to join the community. The community is so involved, in fact, that an emerging problem is accommodating the growing number of patches people want to contribute.

Given this influx, sufficiently reviewing the volume of contributions is becoming a challenge. But when it comes to open source projects like PostgreSQL, a surplus of inspired participation is not a bad problem to have.

Version 10 of PostgreSQL will be available in September. Stay tuned to the VMware Open Source blog and be sure to follow @VMWOpenSource for more updates on VMware’s contributions to the open source community.

The post How PostgreSQL Became an Overnight Success 30 Years in the Making appeared first on Open Source @VMware.

Source: Open Source @VMware