 We are going live and thank you everyone for joining again. This is the OpenShift Commons Briefings Operator Hours today and I'm going to do what I like to say, can you see my screen? Yes. Today we are lucky enough to have with us two gentlemen from Europe representing Percona. We have Roma, the technical director, and Sergei, the product owner from Percona. How are you guys doing this more, or actually I should say, how are you doing this afternoon? I would say it's Sydney, doing awesome, thank you. Sure. And why don't you tell us where you're coming in from? I mentioned you're in Europe, but I think you guys are probably a little bit east of that a little bit, right? That is correct. I'm coming from Moscow, Russia, freezing Moscow, Russia. How about yourself, Roma? Yeah, I'm in Lviv, Ukraine, and luckily we start getting the spring coming, which is unusual for this region at this point in time. You know, we were doing a dry run of this conversation the other day, and I was mentioning to Sergei on the phone, I said, yeah, it's really cold here, and then he was like, yeah, it's really cold here, and I'm in the north of Boston. And I was like, yeah, well, it's nine degrees, and he said, well, you know, it's 12 below zero here, and so he had me beat. Anyways, thanks for thanks for for joining us here later in the day for for your folks. Welcome. And, you know, why don't you tell us a little bit of something about yourselves who'd like to go first, Roma or Sergei? Roma, go first. Okay, so I've joined Perconna like four years ago. Right now, the technical director for Perconna monitoring management, but actually there's a product owner for this team on the project we work in. Technical background, mostly web development before and worked as a web developer to the city of one of the company. So I've jumped into a different side of their IT in general from the development perspective and some management perspective. What what their management needs to know about the products and stuff that the engineer to that's, that's my experience. And where were you at before Perconna? I worked in the biggest marketplace for website templates and services template monster.com. So we create a huge marketplace to sell web designs and all related services to it. Okay, Sergei, how about yourself? What's your what's your claim to fame? Okay, I actually joined Perconna like four or five months ago as a product owner for our Kubernetes operators and cloud initiatives. And I'm fairly technical as well. And before Perconna, I was working at the company called Crossover. It is like a huge remote first company, which is emerging acquisition business and what I was doing there was managing engineering and operations teams. And we were building Kubernetes as a service database as a service VMware as a service. Everything is a service for our internal companies and we were acquiring a lot of them like 120 companies, I think in our portfolio. And hence, this is my interest in Perconna because now we're here at Perconna building DBS databases service on Kubernetes and that is fascinating time to change the world for me. Right on. Okay, well, tell us, tell us about Perconna. I know Perconna has been a household name, household word that's been around I think for since geez at least 2006 right? Yeah, yeah, yeah. Perconna was founded in 2006 exactly by Peter Zaitsev and Vadim Kochenko. Peter is now our CEO. And he's also wearing hats of chief product officer. And Vadim is the CTO. Before that they were working in my my SQL AB. And that is why I guess Perconna is a database company and we're focusing on databases mostly because I want to come from database world. And but what is more important, the corner roots are all about open source. And my SQL AB was great company started my SQL, right? It's used by lots of companies nowadays. And it was like, I believe it's the biggest open source product. All over new, right? And once my SQL AB was acquired by son, my SQL, it started to shift the focus to revenue, changing business models and over promising. And exactly that time Peter and Vadim thought, okay, something is wrong here and we need to start the corner to save the open source values. And Perconna started with Perconna server for my SQL and Perconna server for my SQL nowadays provides my SQL enterprise great functionality, but in open source, right? So it's totally free. You don't need to pay for that at all. So this is, I guess, how Perconna started excellent. And you folks have been working with Red Hat. You have a Red Hat certified operator. I think we've been working with our technical teams. Are you folks, is your Perconna, is that in the Red Hat marketplace as well? Of course, yeah. Yeah, yeah, it is. Yeah, so for for operators. At Perconna, we started this journey for operators, I think, two years ago, we're here in the F. And the way we see operators is just the way to deploy our software in Kubernetes, right? And operators is a way to do that, because it alternates everything for you. And Red Hat marketplace is widely used by the community, by customers, by users, and for us, of course, it's a great place to be. And that is where we put a lot of effort to certify there is a lot of users that we have that rely on our operators. They run it on OpenShift. And yeah, obviously, yeah. Yeah, I was going to say, so is your operator, is that a Helm chart or is that written in Go? It's written in Go. No, it's not a Helm chart. It's written in Go. It's also open source so everyone can use it. And like we have a Helm chart actually, so you can just deploy the operator for a Helm chart if you want. Right. But the operator is more than just deploying, right? I mean, it's for... You know, deploying software is one thing, but managing it in a distributed, you know, Kubernetes environment is a whole other thing, right? Exactly. Yeah, they do operations. This is, I believe, what operators should be solving for users, right? Because deploying is easy, right? You can just roll out your container, but what's you're going to do tomorrow when you need to take it back up or ensure that your software is running in HA. So yeah, this is where operator kicks in. Yeah. And as far as your license goes, is it Apache 2? Is it BSL? What's the licensing model for Percona server? Well, mostly for our products, we use Apache 2. But for example, for Percona monitoring and management, we have AGPL, right Roma? It's licensed under AGPL. Yeah, for far it's AGPL. Yeah, correct. But mostly it is Apache 2, right? By the way, and you know, I think everybody's been on a call where, you know, everyone's video conferencing these days because of these challenging times and here someone's cat rocks across their keyboard or their dogs barking in the background. I let my chickens out this morning, so they're running around in the backyard. So if you hear any strange squawking, it's just the ladies out there stretching their legs. Okay, so let's talk about solving customer challenges these days and my sequel 5.6 and post-UL support. Yeah, this is a big challenge because my sequel 5.6 is, we can call it legacy technology. It's pretty old, right, but it is still used by a lot of people. And even on Amazon, just AOL did just recently. And what we're doing here, because Percona is providing support, consultancy and managed services for customers running their databases. And what we are doing at Percona, we're providing, we continue to provide support for my SQL 5.6 for our customers. And that is quite big because we have a huge customer base running my sequel 5.6. I don't know if it is unfortunate or a good thing, but they certainly need help before they finalize their transition to the next version. Either it's 5.7 or 8.0 right away, but yeah, it's a big step for us. It is a big challenge, but I think there are more to come, right? I mean, lock-in in the cloud. This is a big challenge as well, and we're trying to solve it with our operators and DBS platform. We want to give customers and users free software to run their databases anywhere they want with just a few clicks of the button, like on operators and through an ICU. So this is another big challenge, I believe that we're trying to solve nowadays. And you folks are software agnostic in that you'll provide assistance to customers running any variant of my sequel, you know, MariahDB, MongoDB, Postgres and so forth and so forth, or how does that work? Yes, yes. Roma, do you want to say something as well? Because I'm talking a lot. Yeah, the saying that all versions are supported. So even if the Percona server for MySQL, we still do support for Oracle MySQL version and other things. And really, DBS. And all our software, at least for monitoring supports all versions and all different distributions. Fair enough. Right, so let's talk about CNCF. I know that you wanted to bring this up, Sergei. You had mentioned that there were some challenges with Percona around CNCF landscape. So why don't I tee that up and hand it over to you? Not that I want to bring it up, right? It's like, I just want to talk more about operators. Which is, this is my baby. And yeah, so there are, I would not call it challenges, but running data on Kubernetes is challenging. And there is a community, I don't know if you heard DOK data on Kubernetes community. And I believe it's soon going to be part of, like, huge CNCF community, I don't know how it is called. But the goal of data on Kubernetes is to provide a good and nice, easy way to keep your data safe on Kubernetes. And Percona is, we invest in a lot of time and effort to run your databases safely on Kubernetes. And we have now two operators for MySQL, which is Percona X3DB cluster with synchronous replication and another one for MongoDB. And we have some talks by our team members where we honestly compare the operators from other companies as well. And what we see is that not many companies are investing their time and money into operators nowadays. We see operators appearing like from my ADB. But they are even not saying that, okay, this is our operators production, great, ready. Yeah, right. And we're saying that we have our databases running in production for many customers and we're solving their day-to-day issues. And this is a challenge because operators, and especially databases on Kubernetes is a green field. Right? And keeping your data safe on Kubernetes is the task where you have everything so fMeral. Your reports coming and going, your notes are going up and down all the time. You need to make sure that your data is there, back up, safe and consistent. So, yeah, this is the challenge of building. Okay, well, and I wanted to just remind you folks that we are live on YouTube and Twitch right now. And if people have questions for Roma or Sergey, please feel free. You can put them in the chat and comment window and our producer, Chris Short, will make sure that we get them over to here. So, Roma, you were going to talk about your database as a service. I know that that was a key thing you wanted to talk about here today. Yeah, so with this complexity of Kubernetes and, like, not easy for regular user stuff to deploy and to operate. We came to understanding, okay, but what can we do to simplify people's lives? Because, yes, installing databases, it's not easy work, but work directly with operators, it sometimes also requires additional knowledge. So we start thinking, okay, what's the simplification? And we came to this database as a service solution to let user deploy and create, have databases even simpler, even more in more easiest ways than they can do. So let them configure it and then even develop, so let even developers go somewhere, create databases as they need. So let user to have similar experience as they can have with AWS and other systems, but based on their own infrastructure. And because we had already per corner, we basically combined our operators for Percona server and from MySQL and operators for Percona server for MongoDB and use Percona monitoring management as a UI on top of them. So we're combining our team and Sergey team efforts together and building this tool to simplify people's lives. Just configuration, management and then when it goes to the modification of already running databases, it's also the complex task. So again, we want to simplify this for users. That's where our database as a service solution is coming. Okay, very good. Yeah, yeah, if I can add something, Michael, if you don't mind. So we're not only simplifying the way how people can run the databases is, well, you can use Amazon RDS if you want to simplify running to simply run my Postgres or MySQL, right? But what we're doing more here, our software is open source and you can run it anywhere you want for free. And you can deploy your databases on any other Kubernetes class you want, whether it's in your premises or it's in the cloud, whatever you want. So we can make your deployment hybrid cloud easily. Again, fully open source, right? And what we really want to do here, because we see customers doing that and they're asking us about that, a lot of people don't want to be running on one cloud. A lot of people want to avoid vendor lock-in. And this is where the best from pre-corner will come handy, right? Because it simplifies the way you can run your databases, but also allows you to run them anywhere you want. I have a question with you about databases. You know, I work here at Red Hat, you know, I do partner marketing activities with software companies like yourselves and others. And I talked to a lot of different software companies and specifically database vendors. We have lots and lots of database vendors that certify their software for use on Red Hat OpenShift and Red Hat Enterprise Linux and so forth. What is there a specific type of workload that Percona server is better for? Meaning, like, there's, you know, can you help me figure out what's the difference between, you know, NoSQL, NewSQL, MySQL, you know, companies like you got Couchbase, you know, YugoByte, NuoDB, Percona. Why do we need so many different databases? Okay. Roma, do you want me to take this one? Yeah. Okay, I'll try to answer that one. I believe Roma has the better answer. He knows better. But from my standpoint, if you ask about Percona, why Percona exists, why we have the software, right? It's a fair question, right? And for example, if we take Percona server for MySQL, it has the biggest difference from regular MySQL is that Percona server has all enterprise features, but for free, right? It has it out of the box and you can use it right away instead of paying to Oracle a lot of money. You can use Percona server for MySQL for free. And the same goes for MongoDB. But on top of that, we also have our distributions. And in these distributions, you have great packages that are certified, we can say certified by Percona, which are targeted for DBA usage so that you can keep your databases up at night without any issues, right? This is why we have Percona distributions in the first place, because we ship it with the packages and software which works out of the box, okay? Does it answer your question, Michael? Yeah, that does. Are there different workloads that your product versus some of the other different types, like what's the difference between MySQL, NoSQL, NewSQL? Presumably there must be different workloads that each of those different types of databases address, is that right? Sure, yeah, definitely. Every database is not, well, database like MySQL is not a silver bullet, right? If you want to run huge analytics, it will not work for you. You need to use something else. And there is definitely the wide choice between the database engines. And if you're just asking why there are so many databases out there, well, I can tell you that each product solves its purpose, right? And we can look at the top-use databases, it's still Postgres, it's still MySQL, then it's MongoDB, and then there are more, right? And the reason why there are more, because they serve some go or they deliver some functionality which others don't have, this would be the answer, right? Obviously, you can run everything on MongoDB if you want, or you can run everything in MySQL, but your performance might be poor or your features might not be what you're expecting. Fair enough. Okay, well, you folks are going to show us a technology demo here today, is that correct? Yeah, yeah, yeah, I can share my screen. What I'm going to show is our preview of Divas and how it works, so it will be quick. Can you tell me if you see my screen now? We can see your screen. Okay, you see this is black and this is black, right? Yes. Okay, good, good, good. Okay, so it is the interface of our Preconum Monitoring and Management. It is open source again, it is based on Grafana, Victory Metrics under the hood, and we'll see it's Postgres here, it's all open source products, thankfully packed here. And what I'm going to show you today is our Divas preview. So what it does here, I have a screen where you can add your Kubernetes cluster and you just submit the name and your kubeconfig file and that's it. And once you have the cluster up and running, you can spin up databases there. We have one, but I'm going to create a new one just to show that. So I'll call it on a demo and I'll choose my Kubernetes cluster and the database type MySQL. We also have MongoDB operator as I mentioned, but I don't have MongoDB operator installed on my cluster, so I can only choose MySQL now. And for the sake of simplicity, I'll run single node cluster and you can choose memory, CPU, its request and Kubernetes and the disk size. Disk size is a consistent volume claim, so how much data you want your database to be. And that's it. Then you click. So let's see. I'll run and watch. I'll click create cluster and right here I have my Kubernetes and console connected to that. So you just see now spinning up the pods and they're in the nice UI. You have protesting and once it's active, your database is up and running. But the real value here is not only in spinning up the databases, right? Because we discussed spinning up and deploying the database is quite easy, but it's also managing them. So you can scale them up once it's up. You can edit it. You can view its logs. And also we're going to add the functionality to take manual backups, point in time recovery and everything else. It's also, it's already available in the operator, but it's not yet in the UI. And then the other big thing here, which is the great value of recording monitoring and management is our monitoring itself. Right? Because it was built for that. Once you cluster deploy, you can start monitoring right away without any additional steps. And it's not only CPU and memory monitoring. It provides you a deep analysis of your database itself. So you can 15 minutes. You can look into the basic metrics of extra DB cluster because our operator is the corner extra DB cluster is the corner server for my SQL with Galera and it provides synchronous replication. So you can see basic metrics and detailed metrics of the corner extra DB cluster in Galera. You can, for example, switch to query analytics and look into the detailed output of the queries which are running on your database. And again, it's all working out of the box. So you just click the button. Okay. That's it all. So you just click the button and your database is up and running and your monitoring is up and running as well. So here you see that you can do, you can look into the queries running on this database and details. And I think that cluster is up. If you see the BXC clusters, both of them are ready. So you have everything here and your database is up and running. I believe that's it from the demo side. Do you have any questions, Michael? What about, what about scale? Meaning like, you know, you mentioned that, you know, you can run Percona server, you know, build it once run anywhere on Kubernetes. How well does it scale? You know, if, if, if, for example, a customer wants to do like say, you know, a distributed, can you do a distributed database and multiple different nodes? Yes. Yes, exactly. So it's working out of the blocks like that. I mean, on a single Kubernetes cluster, you have multiple nodes and Percona operators, they rely on NTFINI to review the database pods and different nodes. So we don't have any single point of failure on the database. Was it the question? Yeah. Yeah. Okay. So, and you can easily scale up and down your, your, your database, like from here you can, if you are Percona demo database, click, edit and just put the number of nodes like from one to three. Right. And we'll scale up your database on your Kubernetes cluster from one node to three. And these three nodes will run on different Kubernetes nodes with NTFINI. Okay. What about, what about 10? Well, you can definitely put 10 here, but it is not recommended for PXC, not the operator, not the Kubernetes, but for PXC to run more than five nodes. And one way to be, it won't be a problem you can have, but for PXC as long as it relies on synchronous replication running more than five nodes is kind of challenging. Because it means it needs to write synchronously on all the, I don't know, X10 nodes. It's a lot. Okay. Percona live is your, is your online conference you do it in, in one in Europe, one in the U.S. And I know they, you know, we used to be able to travel around and go to conferences and things in person. Is that all virtual these days? Yeah, that's, that's virtual. And it's the next one is coming in a, in the beginning of May, May 12 and 13. We still have the call for paper open until for next 11 days yet. So, so for next 11 days. So, and that will be Percona live online 2021. Again, as all Percona conferences about the open source databases and and technologies related to that. Last year, Percona live, you released the, the, the open source data management software to survey results. And I was going over those the other day. And, you know, there's lots of companies that put out, you know, survey results, but, you know, certainly, you know, database, probably one of the more core, you know, business apps out there. So, you want to share with us the results of that here. Yeah, yeah, the results already shared on our website. And I would like to highlight some of the, some of the most, I think, important data we got from that result. Because it's showing how the, how the land of open source databases and engineering in general changes because we got like. One, probably some of the results say like, who's decision making on architecture and what's like, what's to use in database, what, how we will build soft and with like 41% of architects responsible for that, not DBA is not management. It's also 26% are developers responsible for what and how the system will be built. So it's, it's continuous shift to, to the developers and like technical people, how to do this, not, not some just managers who decide what, what we will use because of, because of some good white paper. Another point is that 22% of companies said they spend more on the using cloud solutions, cloud vendors, then they have expected so their plans to for cloud was cloud expenses was much lower than, than they've got. And that's where, where we do like additional difference from the corner product from even upstreams we use, we already make and ship them pre optimized for performance. So with rich feature, we also may ensure that we're making what we can do to make in default configuration products running more efficient and give people more options to optimize what they're running. That's, that's why we have like so many metrics in the corner monitoring management, why we, why we care about all the, all the tiny details and date. The, you, Sergey, when we were talking with this the other day, you were, we actually were, were looking at it, do we want to do want to pull that up? I mean, they always say, you know, a picture is worth a thousand words. Sure. Sure. Do you want to do that is you're talking about them. And every, every video conferencing tool is a little different. So we're using blue jeans. So the, the share screen is up at the top. You just have to move your, move your mouse up and it'll pop up. Tell me if you can see my. I can. Yeah. Yeah. Okay. Yeah. So that's, that's the page where we posted all our survey results. That's the, that's the numbers we got. Like one of their interesting stuff is about the, the open source tools. Used. Also there, the growth of data, almost, almost everyone saying that the data they've used is, is growing and that strength probably will not, not stop. The interesting, again, that's the fact related to, to simplicity of database management. People start like previously, like 10 years ago, probably the updating database server was like, oh, that's, that's big and huge deal because you, you need to do a lot of, a lot of stuff. So it's like, you need to prepare the downtime right now. We'll do two or three upgrades of their databases over the year. And there's no way to do this in an old way. It's like, you can do this, but it will be so expensive. So that's where the automation and simplification helps. And again, the, the question about keeping people up at night. It's, as you can see, a lot of people saying it's performance and, and all the problem goes because of it. Yeah, performance, performance is the, is the big concern for lots of respondents. And that is where the corner when you turn in that is it provides you a single pane of glass into your database deployments. So, yeah. I think it, I think it's fair to say that, you know, performance is obviously a top consideration as well as, as well as security. Right. So. I know it's a lot of the vendors that we work with are really focused on, you know, securing those workloads in the cloud because it doesn't matter if it. Doesn't matter if it works. Great. If someone can have access to your, to your database, then you've got serious problems. What was there? Was there a security component to the survey? I'm not sure. Oh, maybe I don't think so. Yes. Security is important, but at least not waking people up at night. Sure. Unexpected stuff. And this, like this huge percent regarding the, the open source. Yeah, it might be related that we are mostly working with people who use open source supported. So we got this, like, what's a big number of people who rely on the open source of saving money. It's I would say it's always fair to say that open source is not free because you still need to invest something, but comparing to some expensive solutions. It's maybe more clear investment and, you know, what you're paying for that's not for some closed box of software. You have the software and you still need someone who can use it. And do you do, do you put out the survey every year? And how do people get involved with it? Yeah, it, it placed on all our communication channels. So we, we do this every year, collect the data and then publish aggregated stuff from different places. We can get people answers. Yeah, I believe we get a lot of answers during corner live events when we ask speakers and attendant people coming to the conference to participate in this survey. Okay. And when does that go out? Was it, did you say November? Yep. Yeah, it works. Well, actually was December. The corner live. And obviously that's going to be virtual again this year or, or you have any plans on, you know, you think it'll be, you think it'll be safe by, by that time. Now, by May, I really doubt it's safe. So yeah, in May, definitely it's going to be offline event. We will see what's in the war in winter. I'm not, I'm not that sure about, you know, KubeCon is in, I think it's November in Los Angeles, right? So I really, really hope that it's going to be offline. Is that what you want to go there? Yep. Sure. I missed last year was in, was supposed to be in the Netherlands, right? It was going to be in the Netherlands and then the whole COVID thing happened and it was just like we started seeing companies shutting down conferences are all around the world and I was just, you know, that I think I still have my ticket for that KubeCon. I don't remember. No, I guess they finally did refund us eventually. But so, so speaking of these challenging times, how are you guys finding, you know, the world that we live in these days? And, you know, and customers, you know, is it is, is everyone, is it getting quiet? Is it getting busy? What's the, what's the net change for you folks? Well, definitely COVID and 2020 had a significant impact in most aspects of our lives. But, you know, some companies were unexpected growth due to offering, due to offering in the metals or services. But I really, really, really, really think that a lot of people want to draw a line 2020 and 2021. In terms of, in terms of business, Proquana is the remote first company and we have, I think, 250 people, everyone working remotely. So we, we didn't have to go through this transition, right into the remote world. So it was simpler for us in terms of customers and revenue. I would say we stayed, we're pretty good here. I mean, we have not seen the therapy impact here. But it's surprising because Proquana is a services company and we provide support, consultancy, managed services and I believe a lot of businesses were impacted here. But it tells a lot that Proquana provides great service and our customers want to retain it. Would you say you got over 240 people in the company? I think it's 250 now. And I don't remember the exact number our HR team shared with us last week that we started to 2020 with, I think, 180 people, right? And then by the end of 2020, we are now 250. And by the end of 2021, I think we're going to be more than 300 something. Well, that's pretty exciting. I know when I started at Red Hat in 2002, and there were 260 people in the whole company worldwide. Yeah, exactly. We once had a company meeting down there in Raleigh, North Carolina, I guess it was on the seacoast there in Wilmington or something. And everybody flew in from around the world. You folks ever have one of those like a global company meeting where you can actually put all the employees in like one spot at a time? Yeah, yeah, we do that. Yeah. Roma, where was the last one? I think it was in some beautiful place, right? Well, we actually, I think for 2020 or 2021, we planned to get to have the whole company meetings because before we had meetings by some groups, we had engineering and a team meeting because it's already was like 60, 70 people. So last meetings were in Croatia and before Portugal, Montenegro, so a lot of different places and there were plans to have the whole company meeting in person. But I think that's some of the biggest, biggest problem we got from the COVID because the remote first culture of company led us almost not get any impact on the COVID stuff. Okay. Yeah. Well, so what about predictions? Not about COVID and when the vaccines are going to be rolling out, but you know, where do you think computing is going to be 18 to 24 months? Are we there? Is Kubernetes it? Or is it, is Kubernetes going to become the next open stack and then there'll be some new shiny object that'll be coming down the pipe? Your base is going to become a new open stack in terms that it's going to be not that widely used anymore. Is that what you're asking? You remember, it was 2000 and whenever like, you know, the open stack, you know, conferences were like really huge and it was just like open stack this open stack that neutron and Cinder and everybody was on board. This is the next generation thing and now it, you know, no offense to open stack, but it seems to be, you know, very niche focused in the telco space and with some very, very large customers and now it's Kubernetes. So, you know, is this it? Is this end game or, you know, what are your predictions for 24 months from now? Will be, you know, is KubeCon going to be 50,000 people or is it going to be, is going to be niche? No, I believe in terms of if we talk about the predictions in terms of Kubernetes, I see Kubernetes now and I think, I think Brandon Bernstahl that that Kubernetes now is a MS-DOS, right? And slowly it's going to evolve into Windows 95 and then to Windows 10 and then something greater. So there is a trend like you can see recent announcement of Google GKE Autopilot, right? So there is a trend to making Kubernetes, and I believe in 24 months we're going to, in this period from today to 24 months, we're going to see a lot more announcements like that, which are going to extract Kubernetes from developers and everyone, right? So you're going to just run your workloads in a stateless mode. But I believe Kubernetes is going to be the underlying infrastructure for more workloads in your future. In terms of other predictions, we had good, we have HOS, it is Matt Yankovich, he is our open source guy. He does great predictions for every year and for 2021, we have three bold predictions. And I can quickly go through them that one of them was that DBA role and regular system means they're going to, not Danish, but they're going to be merged with other roles like as a reason developers. So there will be no more like a dedicated DBA, for example. We're going to have an S3 slash developer who's going to run databases because databases are simplifying more and more. And the day two operations are now easier. So you don't need to manage your underlying hardware for your database. For example, you need to manage your database itself like wireless and so on. And you don't need a dedicated system administrator to do that. It's number one, which is quite interesting, I believe it's valid. Another one is that open source software is one. That is quite interesting in the sense that we see that more and more projects and companies are adopting open source software or maybe there is components like they take MongoDB and put it somewhere. And on top of that, they build their own DBAs. So why not? But at the same time, we see that big companies like Amazon, Microsoft, they invest a lot into open source. And while open source vendors, traditional open source vendors like MongoDB Elasticsearch or Radis, they are introducing more restrictive licenses for their software. So it's kind of a shift to the enterprises investing a lot into open source and like cooled old fashioned open source companies. They are closing their source of changing the licenses and it is interesting what is happening nowadays. And at the corner, our motto is to keep open source open and we want to continue doing that anyway. And the third big prediction is that consumers and users will revolt against this open source in any guns. So you can see how it turned out for Docker, right? They started playing these games where they locked people out and limiting their downloads of images from Docker Hub. And it turned out that a lot of people decided, okay, I want to move to Kauai, for example, where I will run my own registry. And yeah, and this might keep the community and other companies apart. I mean, if open source is changing their model, people will not believe in open source text anymore and they will turn towards enterprises or other companies. I was going to say that was pretty surprising. I mean, Docker was the biggest thing since sliced bread, you know, four years ago. I mean, obviously Docker, the technology is still mainstream. But given that so many people from Red Hat left Red Hat to go over to Docker, it's amazing to see how they couldn't keep it together. Not that not that they're gone, but it certainly didn't play out the way that I think they were expecting it to. Yeah, yeah, I agree. Yeah, it's interesting. Okay, well, we've got we've got just a couple of minutes left here. If, you know, if your head of marketing for Percona was here on the call, they would make sure they'd want to say one thing. What might be that one thing that they want to share here while we're on the call today? I believe not only the head of marketing, but everyone at Percona, say just one thing, keep open source open. This is this is very important for for our company and community. We're working really hard to to introduce traction to make community work together to contribute to make free products and keep them open open source and free. Okay. Over to you, Roma. Hello from you. Well, I would say that they continue like in addition to open source is like, we are trying to and like, I hope, successfully avoid any vendor looking for people who use Percona product. So, with any product, you can switch from database for the server from my SQL, you can use my SQL server from Oracle, you can switch to Percona server, you can switch back. It's if you're not used to like really specific shifts, it will work. So, we're doing everything we can to do not lock people because that's our. Okay, fair enough. Well, thanks for coming on today really appreciate you folks being here. I'm going to going to do what I like to say, I'm going to share my screen. And can you see my screen? Yes. Yeah. Right. So, Q and a what's next calls to action. Make sure you check them out. You know, on GitHub. Head over to their to their main website and learn about their Kubernetes operators. And I noticed that you didn't put your home phone numbers up here. Sergey. It's just just email addresses. Well, it's, it's well known for everyone's great. Well, thanks for coming today. I hope we can have you back and you know, you should catch up with our team. If you guys want to be part of our podcast series. I think I think we'd love to have you be join us doing 1 of our Red Hat X podcast as well. So. Thank you very much, Michael for having us as well. Thank you. Enjoy the, enjoy the weather over there. I hope the snow melts sooner rather than later over there in Moscow. Yeah, likewise, likewise. But as you see, I also play hockey so we need that snow. Right on. Thank you. I guess again, we'll talk to you later. Yeah. Bye.