PostgreSQL Community and Why PostgreSQL Continues to Grow in Popularity - Percona Podcast 40

by Umair Shahid, Matt Yonkovit

Link to listen and subscribe: PodBean

The Percona HOSS, Matt Yonkovit sits down with Umair Shahid, Head of PostgreSQL at Percona, to chat about his background, the PostgreSQL community, and why PostgreSQL continues to grow in popularity. Umair counts 20 years experience in PostgreSQL and open source experience serving some of the highly competitive technology markets in USA & Europe in roles of increasing responsibility.



Umair Shahid

Head of PostgreSQL at Percona

Umair counts 20 years experience in PostgreSQL and open source experience serving some of the highly competitive technology markets in USA & Europe in roles of increasing responsibility. Proven track record of the unique ability to bridge the gap between business objectives and engineering :

  • Product management, with a focus on growth
  • Data-driven business development
  • Leadership skills with the ability to rally a geographically dispersed team

See all talks by Umair Shahid »

Matt Yonkovit

The HOSS, Percona

Matt is currently working as the Head of Open Source Strategy (HOSS) for Percona, a leader in open source database software and services. He has over 15 years of experience in the open source industry including over 10 years of executive-level experience leading open source teams. Matt’s experience merges the technical and business aspects of the open source database experience with both a passion for hands on development and management and the leadership of building strong teams. During his time he has created or managed business units responsible for service delivery ( consulting, support, and managed services ), customer success, product management, marketing, and operations. He currently leads efforts around Percona’s OSPO, community, and developer relations efforts. He hosts the HOSS talks FOSS podcast, writes regularly, and shares his MySQL and PostgreSQL knowledge as often as possible.

See all talks by Matt Yonkovit »


Matt Yonkovit: Everyone, welcome to another HOSS Talks FOSS. Today I’m here with Umair Shahid, from Percona. In the Postgres community, longtime member of the Postgres community. You’ve been working with Postgres related companies, since 2003. Maybe tell us a little bit about your background here

Umair Shahid: Well, that’s interesting. By the way, thank you for having me here. It’s interesting that the start of my Postgres career was, well, not by choice, I was kind of pushed into it. So we were a team of developers that were working in a very typical offshore software development model, where you have this you know, company in the USA that has this offshore office that can provide low-cost engineering expertise. And we were able to deliver some really good software, we were able to deliver some really good engineering expertise. That and one of them, one of the partners of the company decided to branch out, get into products. And the idea was to take an open-source database and try and build Oracle compatibility on top of it, in order to ease the migration of people trying to move away from Oracle and onto the open-source. And he bought out the office in Islamabad, where I was working, this is the engineering team that I want to build a database with. It just so turned out that the database was that we selected was Postgres and the company that eventually was created to offer that that product was EDB.

Matt Yonkovit: There you go. And so that led you to be the Director of Development Services at EDB

Umair Shahid: Yes, so I started my journey there, as you know, as the developer of their ETL product, the idea was to help automate the transfer of data from Oracle to Postgres, that the first version of the ETL was something that I wrote, eventually, I ended up being responsible for QA, for build, and then being the Director of Development Services.

Matt Yonkovit: Okay, and I mean, from there, you could just continued on I mean you spent some time at OpenSCG, which eventually got bought by Amazon, and then the Second Quadrant as well. So you’ve, you’ve had a very diverse background in the Postgres space.

Umair Shahid: Yeah, and I think that I’ve been blessed a bit, that I was lucky to be exposed that way and to be able to be a part of that community. And work with this pretty phenomenal project. Postgres at OpenSCG, it was all about consulting very hands-on, on customer site. So I was moving around quite a bit. And that deity of focus with business intelligence, creating data warehouses, data Mart’s, ETLs using Postgres, allowing big companies to handle lots and lots of data, huge volumes, huge velocities, and be entered try and make sense out of it. And then but a second quarter, did the roll changed again, in again it was a, it started off as a services company that evolved into a software provider. And I was brought in to help with that evolution. So it was indeed pretty, pretty diverse. Using the same technology, using the same project using the same database. I was exposed to quite a lot of different aspects of it.

Matt Yonkovit: Yeah, and it’s interesting I normally ask people what drew them to a specific technology or software. Obviously, you mentioned that you were kind of forced into it, but why did you stay with it? Like what got you hooked. So you’ve spent your, almost your entire career working in Postgres.

Umair Shahid: I think I think it’s just the initial part where I would say that I was kind of forced into it. I think force is a strong word. It just took it just happened. It was but I think eventually I just fell in love with the technology. I became very comfortable with the community. I felt accepted and I felt that the community was very open to allowing new entrance. And you know, the the the project itself was growing at phenomenal speed. People moving away from other databases towards Postgres features were being added every year with every major release, and just the profit typically being made by leaps and bounds. I got hooked.

Matt Yonkovit: Now, you mentioned the community and one of the unique things about Postgres, that is, it’s really unique amongst open source databases, if you think about it, there is no central owner. And you know, the licence is very open. And so that means that anybody can use it, and there are several companies in the ecosystem. But that lack of an owner if you will, is very unique, when you look at the open-source database ecosystem that’s out there. You see the MySQL is you see the Mongo, well, if you call Mongo open source anymore, but you see the various other databases that are out there, whether it’s Clickhouse or do all of these different technologies. And they typically have a driving company, who is kind of the owner of the IP, even if it is an open-source technology. Suppose Chris’s stands out as purer when it comes to that more open-source movement, then then a lot of the other ones. And that’s not to say the other ones are wrong, it’s different. And how does this work? You know, tell us a little bit about that background, tell us a little like how that uniqueness works, because you’ve got such a diverse audience working on.

Umair Shahid:
Now, I think it comes with its own challenges. So it’s not all glory, and it’s not all there are definitely challenges in there. I once wrote a series of blogs on the reasons why I love Postgres, and one of those blogs addressed the topic of what you just the question that you just asked, and he said, it cannot be bought out, there is no entity that’s driving the community, the project. There’s no single company that owns the project, or the intellectual property that can be bought out. And I think that makes Postgres very, very unique. It’s a pure open source project. There aren’t forks, there are products built on Postgres, but the project itself is, well, as open-source as it gets, the licence is as liberal as it gets. And I cared about this, I said that the only thing that licences in that text is, please don’t blame the University of California if something goes wrong. That’s the only limitation. But yeah, I think that at times, it feels that there, there is a lot of progress being made, because there are so many diverse people pitching into the project. And with no single shepherd in place. You know, it’s driven by consensus. And sometimes that consensus becomes hard to achieve, sometimes, and consensus takes very long to achieve. But eventually, what goes in makes for a product that’s robust, that’s reliable. And that’s been growing in terms of features. And in terms of functionality and popularity. JIRA was here, consistently over the past 20 odd years.

Matt Yonkovit: So now, you mentioned the popularity and Postgres has had this interesting journey over 20 years. In fact, I use Postgres, long, long ago and then came back to it more recently, so there wasn’t there was definitely a gap. But I remember back in the day, I think it was Postgres 6 or something around there that I started so it was, it was quite some time ago. But over the course of either the time when I first looked at it for 10 years, it wasn’t always in the conversation when it came to databases in an enterprise. In fact, it probably was left out more than other databases. But over the last five years, that’s changed. In fact, over the last three, four years, that’s changed dramatically. In fact, Postgres so Red Hat, you look at you know, Stack Overflow or JetBrains just did the survey. And you know, you’ve got this insane growth in terms of popularity in terms of people viewing Postgres with different eyes. In fact, Postgres is StackOverflow, fastest-growing, most popular and most loved database. I mean so I mean, and they’ve got, like, 80,000 people that they survey in the developer community. So that growth, it’s always been there steady, but all of a sudden, the last few years, there’s just skyrocketed. What Why? Why is that? I mean, like, what’s driving that?

Umair Shahid:
I think it may be very hard to point to a single element that may be driving it, I think it’s, it’s a combination. And when I first started working with Postgres and us, and this is something that I relate to people that talk to them about Postgres, when I first started working with it, people used to opt for it in order to reduce their costs, that was a primary driving factor. While that factor remains today, the total cost of ownership is significantly lower with Postgres than other databases or competitive databases. I would say that that’s not the only reason that may not even be the primary reason anymore. If that year, over year, the addition of features that Postgres continues to have, that’s primarily driven by demand offered to users. So the demand comes in, the future gets incorporated, it becomes popular, adoption increases, and that circle goes on that’s definitely one element to it. And then I would also say that, as the ecosystem matures itself, as various commercial companies start to build a business around the project, they start pumping money into the promotion, and into the development of features and tools to use the database. And I think that is a big contributing factor as well. So when commercial companies employ some of these, these contributors, when they help finance, various different conferences, so that people can come together and collaborate and discuss when they fund people who are developing some of those features that would otherwise have not been developed, or at least not as rapid, that, that that impacts things. And then maybe there’s a third angle to it as well, that may be the environment, and the fact that some of the other databases that may not be as pure open source, as Postgres is, are maybe driven by certain corporate interests. As, as they evolve. And as corporate interest or vested interest come into play into the development efforts, and into licencing. People get put off, and they choose more liberal licences, more liberal communities. And that may be you know, that may be another reason why Postgres becomes more popular over time. So, I guess it’s the, it’s the acceleration of all different factors combined, that contributes to, to popularity.

Matt Yonkovit: So now, you mentioned that the the money coming in the contributions from different companies, and it’s not just code contribution, sometimes it’s the sponsorships of the conferences, sometimes it’s the paying for developers to work on the core and contribute back code. Sometimes it’s other efforts. But I believe there are somewhere around 7,292.3 different versions of Postgres available. I exaggerate that number a little bit. But it seems like there are so many different versions of Postgres out in the ecosystem. And it’s, it’s so many, and how do you navigate this because everyone seems to have their own version?

Umair Shahid: Well, I think that that partially is driven by the enterprise need to have a certified version from a certain company that is driven by SLA s and by guarantees that the community does not offer a guarantee. The community is well open-source, and the only thing that they say is, don’t blame the University of California. So, so if, if I’m an enterprise, and I’m investing in, say, 5000 servers that would be running this database that’s going to, I don’t know, handle a trillion dollars. Have transactions every month, I need some guarantees, I need somebody. To put it in a blunt way, I need somebody to sue in case something goes wrong. Okay, so, so experienced this in the past that even though I would try and promote open-source Postgres, saying that you’ve got every bit of functionality that you need, right here, you don’t need a proprietary fork, you don’t need a proprietary version, some of these enterprises are looking for the commercial offering, even if it says, if it is a simple wrapper that puts a stamp, a company stamp on top of it, saying that this is certified by company x, they need the certification, they need some kind of a contract to go with, in order to be able to seriously use the database in production. That would be one reason why there are many versions out there, they may not deviate too much from upstream. And the ones that do deviate, have such a major overhead and trying to maintain the fork that everybody that I know, who have tried that regretted doing it?

Matt Yonkovit: Well, and I mean it’s not just software companies, it’s cloud providers and everything else, all of these, all of these different providers out there, have their own fork or their own deviation, without a branch, if you want, if you don’t want to call it a pure fork, where they’ve integrated with maybe their, their cloud infrastructure, maybe they’ve integrated with some third party things, maybe they’ve integrated with other services they offer. But it changes, what is the core, which makes it somewhat compatible, but not always. And when we talk about some of the enterprise versions that are out there, I think that you know, and you tell me, but sometimes moving towards an enterprise version that is more of a fork versus you know, it’s not deviated as much, sometimes it makes it harder to go back. Because you get locked into that version, or that that feature set or so. So is backward compatibility is something that a lot of the forks are looking at.

Umair Shahid: I think it depends on the fork that we’re discussing, or that we’re talking about, definitely are folks out there that have deviated so much from upstream, that there is no backwards compatibility. You know, there’s no on-disk compatibility with open-source Postgres. And that that makes us makes it hard to go back. And it’s, it’s sad that I find it sad that you lean on an open-source offering, and you claim to be an open-source database. And yet, the very thing that open source tries to avoid and that’s vendor lock-in customers still walking into it, and sometimes not even knowing that they’re walking into it. And I think that me personally I’ve tried to stand for open source Postgres, even if it is a simple certification of a company that’s come in, maintain on-disk compatibility, so that, even if you know, you are going with a certified version, if you want to move to someplace else, you can easily just pick up your data and go there. But that’s not true for all forks, I think I think it would depend on what you’re talking about. You know, I can talk about Percona, if you know, if I may represent Percona in this conversation, and I know that this is something that we, that we cherish, and we hold near and dear to ourselves in keeping that on-disk compatibility, keeping that open-source compatibility so that there’s never the chance of a vendor lock-in.

Matt Yonkovit: Okay. Okay. Umair, thank you for stopping by today and chatting with us. I appreciate the time you’ve spent with any last words for us?

Umair Shahid: Oh, well, I guess Postgres continues to continue to grow in popularity. And I hope that we can all come together to try and identify what might be factors that people need addressed in that database and address them and make it even more popular.

Matt Yonkovit:
Wow, what a great episode that was, we really appreciate you coming and checking it out. We hope that you love open source as much as we do. If you like this video, go ahead and subscribe to us on the YouTube channel. Follow us on Facebook, Twitter, Instagram and LinkedIn. And of course, tune into next week’s episode. We really appreciate you coming and talking open-source with us.

Did you like this post? Why not read more?

✎ Edit this page on GitHub