 okay so you do have the option to yeah hold on let me just see if that work it's flickering so just keep it on camera this is 1080i okay it's good yeah still flickering that's fine keep it on camera for now I'll just go and check it but yeah yep on two I'll zoom out a little bit yeah can you I can speak into this for this is not good go ahead check check check right here so that tells you that it's yeah so it's fine you so you're gonna be a check in one two hey hey hey that one you know you still got me in so you see that one oh she's fine yep you can have that if I like this one customer this song loud one two one two thanks for joining us I'm Preston Wade and I'm here to talk to you about the marketing and open resource software development in cloud like it's ready and I'm trying to see the subtitle I think we're going to talk about why operations matter, what we're actually providing here on why someone wanted to make a difference. I'm going to talk about operations, what that process is, I'm not so convinced about using it and why they're doing it so we can start using inspirations then the next chunk of time is going to be, the person is going to talk in the next steps of how the service is onboarded, so how the software project would be included. So all that needs to be done. So first just a little bit about our reports. So what we're doing here is we're building an all open source community cloud planning because we want to learn back how to make software and all the cloud operations that's better by running those services and then open operations. So it's sort of normal open source development, which is the point you need to say now because two years ago there wasn't such a thing, there wasn't being able to make it normal. In a regular process this is how to change those, but in a stream development process and that goes into testing environment and integration and on the other side of that you can do deployment and what is the difference. But it goes to users. At that point in time those users can then use, in that whole procedure, experimental software or at least software that can be directly connected to that stream and that will make improvements in the future. That's sort of the power of open source software in a lot of ways. So what we're doing here with the awkward person methodology is essentially the same process in terms of studying the stream development or in the same testing process but then there's another step after the step CD and that is to bring you to the cloud environment. That is a problem, a whole production, live environment with all the other software that you're running for. It's not a testing system that's brought up, tested and done right now. It's an onboarding environment to be running in. And then it's delivered to your users. Well it actually gets another layer of testing, another layer of improvements and what we learn in it can be find it, find it to go back. This is all for yourself. People and users can write an article and see that and so forth. So of course users get their regular chain of feedback to the developers and I'm not sure that many cases of chance to get to the software faster because they're not having to install some of these cloud services in the future, seeing how that future is working. And in addition, it gives them an opportunity to test and prove that software in their own environments while it's still being developed so they can quite feedback on proving it and they can provide the other software in terms of the day or the surface or whatever. This is a change that they're seeing. Your environment is all going to be as well. And at the same time, other of the sort of projects that are involved in the software are in there as well. They're learning and improving, perhaps, and knowledge. So that community cloud space becomes a creative practice and an actual place where things can be done and we can bring out for services. You want to think about what's the surface level expectation is like an output service, and you're not trying to compete to a virtual environment. So in order to put this together, this is from this offering first idea of the two things that we have said and all of that, we put together a project, an open source project that's set up by opening operations. And I have, this is open sourcing operations. I'm trying to change the language because open source is not a very, you know, what we're advising you now but opening is a very, I mean, that kind of opening, especially, so you'll see that kind of throughout. But essentially, that means that we, you know, we're saying if you run your software, paint production before you deliver, and then you build a prototype of operating in first environment, and then you have an operating dashboard that will operate for us if you can talk about it. I think it will work as well. I think you might have seen a lot of stuff that will keep the same results. And then we kind of do that because it's going to talk about a concept but we want to actually show that your idea is going to do something. It's like we're saying, come test me with the software. We're testing, we're testing, including a concept with everyone else. And that's part of the information here to do is, you know, looking at open source projects that you want to do. And so we're ready, we're now prepared to, especially with kind of a high touch at this point in time, I'm sure you could bring a project support. Of course we want, that's not ideal. It should be much more self-service. And that's what we're going to do. We don't have that because we're not an existing cloud domain, a commercial cloud that has all this stuff there for the last two years. This is just that we're all developing together to be open to having this system. And the same thing with SRA practices, is that we are here. The development is in the open, we're writing, we're providing a lot of participation from the necessary teams to deliver conceptual documents, our PBRs, architectural decision records, and so forth, who prints on how to do things. And you can take and create your own cloud based on your own. And so recently, recently we've been working on this kind of message in front of the game. That's what you can have an idea about having some set of principles. It's quite the first audience that's actually seen this. I think next week, this is our last week, these four principles. And so a lot of words we're already getting across from the room. I think it's just the first time it's been out, but I don't want to just read a slide full, but I'll go ahead and do that and add to it. So I'm going to first look into the concepts. Relying on standing definitions of open and open source and open one. That's what I'm saying before about the opens, right? So the keys are operate, open, governance, and hybrid. Those are the four principles. By that we mean operate and build software, optimize and run as a service, and allow others to run it as a service easily, including document how we operate it, how we operate the cloud and run itself. Open. We develop open source practices, tooling and documentation and actual software content and everything that allows us to build and manage our software. The corporate lessons we've learned from operations back in the code automate where possible in each of these open source communities. Governments. And this is an important part is that something that's always the same when there's those cloud environment fields of governance and anything like that. So we've worked on fostering an inclusive, respectful and collaborative community through clear standards and accepted practices. For example, codes of conduct, dispenser, and power. And the last one is that it's hybrid. And this is being another a number of measures on that. We ensure that there's no dependency actually to the cloud environment we're operating in. I always bear in mind that software and management works identically on at least one other platform, one other cloud or other kind of business. So these are as directional. Our approach to that environment is mostly like this. We're just trying to be our conceptual investment. So just a quick little bit about what's been happening in other versions of using this stuff. OS Clidement is an important example of this. They're an open source collaboration community and they're building data and software platforms around modeling, financial markets, and investments to consider climate change. Because all of them also consider climate change. So this is a, what the community of practice is all about. So this is a, what the community of practice and development around data science that's happening. And they're creating a group of data commons of all of the models that they've got and then creating and providing some different analysis around with each other. And so those workloads, so the workloads are running on top of Open Data Hub, which is the upstream of Red Hat within the data science product. And so this team has been doing this for the last, most of the last year. So it's providing the ability to quickly scale up and bring in members of the community to work directly as a developer to ask data scientists without having to learn a whole bunch of human and whole bunch of processes. And it allows them to work on, to work from idea all the way through innovation all the way through the struggle, all the way through the competition. So yeah, so essentially I'm just looking to see how these work really go in software operations match up. I mean this is another way of looking at the aspects of this for principles for the work. But in a sense, it's a lot of them to come up with on-time open source communities. We're not spending a lot of time making decisions around which tools we use and who's going to pay for the AWS bail. And so I wanted to think of it as an approach that I'm working on. So I'm just going to give you the next part of this and start doing some of it. So as with all the different components, it's for something creating workloads, like this each project of data process that you can interpret in different scenes. And this is another thing that we also work running management optimization on web first. The example here ACMI is the main problem. And we also use it in the service. We check our data from web first steps. Because of what you can see on the server. The project of our content besides the current aspects of it. First we're trying to build a government score on the source code so we can also build their kind of group friends. It's an example of a really interesting problem that we've been working on here for the last time already. Part of the model here is that we provide open telemetry. We did obligate it to be with the DII, the kind of information that's invented for our organization. So people are using it on their power to scrub and anonymize data. So these are problems that people have that actually had a reason that's why I think this would be that problem. So what is all this concept behind where can I start? For this, all these different workloads and the service you have you're kind of improving all kind of personas. These two examples I'm focusing on the developer. Two examples to make oneness and of course communication as a whole. The rules as a service is already something on it. So let's talk about ineness using githubs to manage the audience and the stores for people to work on. Then you have this dynamic file that you can set by when you can first is the rules as a service. So instead of running your own by this and then you can search and this is what you can do in the program. So it's great it's down that's a positive problem in that way by itself the service code the thing I want to point out is that unfortunately the rule that was just set in there in terms of how that is going to be beneficial for you it's unfortunately not. One of the things that's important to us is a lot of people don't know what a hub is and when they hear that it makes it good. So today our documentation on our process is a lot to teach and learn from people working on improving the documentation of the goal is to make it so that data scientists can come in and they can not have to worry about any of the problems. So there's a gap between the intention that we are able to put to that all the time. So I want to encourage you then that you felt like this doesn't sound like you but there's a barrier coming. That's it before we do an extra high touch right now because there's something that's improved on HM and automation before that happens. You point to this one because the next example and she actually wrote an amazing blog on it and told me what it is and what it is about. So she used data to get a kind of magic data on this case. So this is a basic idea the difference between data can relate to all of these which you guys state in the same way and you can just put it in there. So why this example is happening when you look for stars because you will go in this direction and you can do a quick turn to get some data even better the ideas are already in the very large amount of data and you can do all this with a small draw and in short time in this case I'm just going for the steps. So she made it really easy by how many to get it and how to set up to a yellow silence and short with the same day in space and and when you open them data will just be connected to this and she wrote this in her little repository we wrote all the different files in the database in there and at the end we can just easily go to our first last and smart and find all the different settings in the different files we will see how easy you can connect the two lines quote and it's going to be very invisible we are not going to have stuff here because it's some other data science yesterday but an easy point easy follow the different web pages of the data that is collected here and follow the what file and you can 20 minutes that's really short if you want to follow the link the page is also online so just go for the steps we'll be running in a few minutes and you'll all get your work done and this is at the end the link page or anything that you just did follow our Twitter and most of the topics I'm recommending going to our repository because there's a lot of stuff happening so please sign up and get all the help from folks here and so there are a lot of recorded presentations out there right now our YouTube channel has a number of pieces going on the things that people that could see exactly so we chose not to get into that today but we didn't have to fight for this because we didn't have anything with any allocations but what's nice about using the blog is a lot of people this is how documentation gets created to start with and to start with one account this gives you an open source project with a mindset about what you want to try out something to walk through and model if you write down some steps and help us get that published then we'll be happy to do that we have wanted to that's our presentation one to do a couple minutes less questions we've got a mic in the back but I need to do something to stick the mic up but then you can still clear yes, we can still clear thanks I'm a big fan of Operate First and I'm really glad to see all the open source communities projects kind of seem to start to aggregate on your community cloud I'm very interested in the distributed telemetry do you have any stories of what people have been able to discover about how their services are running by using the distributed telemetry or even just are there any astonishing discoveries when they actually try to run their software on a real cloud do you think that's mine for you or one of the things one of the things it's a big question I think that's something we're trying to track right now with the stories of people about what's going on so I know what gets in and then it doesn't just remind me all this kind of story because there have been several things that have come from what's going on in the open grid about what's going on in the open grid the way that products are going to be and part of this is that we don't have a full telemetry staff at the company running so we can then go back for that yeah these are first steps the kinds of gaps that we're getting from people right now what you thought it was going to take to get your service running and your knowledge is coming back this question so this is very interesting to me as someone who's had to deal with applications and had to figure out how they're supposed to be deployed because nobody else does including the app developers themselves especially with a lot of open source projects where people are doing this sort of without a native kind of approach of developing applications I mean you guys started figuring out outreach approaches to getting open source projects to start looking at as part of their development process as part of them actually making their applications and whatever to consider first coming to Opry first and build their applications with the full life cycle of an application on the process in fact we're experiencing that right now so I started working on the project here and it's immediately with the possibilities but I was in my experience with open source projects that you're saying hey you've got a cool platform you can use it and it's not necessarily so what we focused on was and also because it was going to take those journeys to know so we focused on what happens so we knew we could ask this out of the game and do that so now this is the first time we were starting with the OO and certainly we're beginning outreach because we wanted to get a point where and then also making it really clear that this isn't just like an easy to use free resource that you can work with there's a lot of things to use with it there's going to be some steps involved there's a pain involved with OO that will be going on in the future and I think in essence from the we talked about augmentation making sure that when you come in, at least a lot of people come in and just don't count what you say and what you say and just make a PR and that's like it's not, we make sure that we're very careful about it so the impression I got from having to work with you know random open source projects that open stack or organize whatever is that they don't actually know how you're supposed to run it and the works on my machine gets real hurtful right at that point because they don't really have things like practicing how to run it instrumented things in that nature so that's something that I would like to see come out as a positive result of this effort and that part of those defined like the conceptual answers that you need to have and actually make the creative choices that are providing to you both from your practice Thanks very much for your presentation two quick ones so it wasn't quite clear to me the intention was to search for applications for a long term or if they had a particular maturity point that they were intended to all for stand up their own structure oh it's a good question but it's like a right idea that's a little bit different from my perspective we're not we're asking we're providing resources for open source projects at that point like we all need workloads if you bring your software here that software needs to be running and testing something like I'm not going to go over it from users to your software I can see a lot of it's like we're not going to we're not going to take the likes that opens off per se but because it's more of an and I think what we're now that I'm seeing this out loud I'm thinking that that's probably the part that's maybe that's something that we've added to the onboarding processes like the discussion with the project and creating a set of these ones that we know what's going to happen with those ones and we keep developing a service that we're adding to what we're going to be writing in it could run forever and I think it's just I was just interested if there was a particular maturity from our data but that's a good thing we might find a thing of experience and just a second one for me just around maturity and observability I was just looking at it it looks like you have like a sheer confanance as I assume that any of the service maintainers have access to ideal to find your own dashboards and have matrices is that the case? I have a picture request out on this is that an issue a project page so there's two aspects that we want to see one is the front of what the world sees the one in the world comes and they'll say where the project is how it's constructed what are the set of things that the project wants to show to the world and then there's the dashboard that are there to develop and I think we're just starting to get some of the some of the clusters that are showing I guess I touch on the community currently but now I can write on the on the website but yeah there's a difference between what we want to provide and where we're at right now so yeah, that's the do you help them to find the rest of those and things like that or at least maybe solve to order best practices around finding those great question that you get so for example we're we have a member of our community it's essentially the it's the name of the research cloud that is going life as life as well to be students and that and so that is in operations there's been many school activities that go on time so we're going to be working in a long term discussion about an SLO process with that so not like the other stuff I'm writing and then we're getting in and as you show up where I'm basically thinking to point it out some of the contributions that we get from the SRE team that Brett had has been around finding SLO so there's now a full SLO and scrap guide to use when you want to have those discussions to work with a team so we're using that to have those discussions with that team the first one at that level that's a big piece and then I think now that I've got that framework we're going to be adding to be a a lighter weight version of a a difference of projects in so much kind of a better way of discussion so then so I can kind of iterate on so the goal is to be able to provide what doesn't exist in the world or the best practice that's out there so we've written up that's perfect otherwise as it turns out there's not a lot of open source open content around SRE practices appreciate it, Brett Holmes, thank you that's another hand over here a little bit of my question which was I was wondering about some of the origins and some of the things that you were taking into direct inspiration how are you trying to do more architecture and the best practices and I think a really important question was we're going to open it for the last project and that's part of the dynamic except for the cloud people, cloud native for the people who started some of the work part of the open sector is a part of the concept of running open operations is something that has been going on for a while there's a number of the entire cloud foundation it all runs open infrastructure that essentially everything but the passwords is available and the order of business as well there's a lot of projects to do and so it's a model that's been out there for a while this idea of taking something up to the cloud level I feel like it's kind of a natural next step that's going to happen and then just starting to get out to give it to you how to actually preview the cloud and then to the next level I feel like that's very important there is a sign down there for the closest operation yeah and I think if some of these people are interested in working on this we can do the work daily of course on the cloud practice level but also in terms of like the disruption of the industry you can kind of expect that something like this is going to happen all the private cloud providers have the same level of disruption on this right so and this is how we all do it and so we definitely have time if there's any other questions if we're hanging out here I'd be happy to take any more one observation is interesting one choice for operators first it's pushing and the other one is like I work with Red Hat SSO and we were very interested in working with SSO so someone probably will reach out and the other thing is like how is this different from RH OBS Red Hat for Observer so that is a product I don't actually know the details of the product but the I think one of the aspects is that one thing is that it's probably different for RBS is that we're running bare metal we're running an open chip on bare metal we're running some bare metal and that's something we need for that for other services on bare metal that's what we want to do that's what we want to maintain so we definitely have access to be able to provide more deeper access to abstracting of please much about the potentials and then wherever the components are that would be either they are already installed or they could be installed because it fits everybody