 Hello everyone. Good morning, afternoon, evening, night, whatever is in your time zone or as one of my friends always says, good day. And one day when we will broadcast this show to other planets as well, we will come up with another word. But until then, good day everyone. You are watching Open Infra Live right now that is brought to you by the Open Infrastructure Foundation. We are coming to you every Thursday at 1400 UTC with a little bit of a test how quick you can convert UTC into your own time zone. And our episodes are featuring a lot of good content, production case studies, open source demos, conversations with industry experts like the one today and also updates from the global open infrastructure community. And your story can be one of these. So also if you have any topics that you think would be great on an upcoming Open Infra Live episode, please throw them in ideas.openinfra.live so we can pick them up and feature you on the show as well. Also if you would happen to miss a live episode, no stress, you can go to YouTube and check out our playlist and watch all previous episodes. Like I would like to point out the one from last week where we talked about job openings and opportunities in the open source ecosystem. So if you're interested in that, go after this episode and check that one out. Today this is the 12th episode. My name is Ildiko Vanca. I will be your host for today. And I'm joined with two experts here, Paul Miller from Wind River and Mark Collier from the Open Infra Foundation. And we have a very hot topic for you today. Some of you may even consider it as a hot potato by now. Yes, we will be talking about Edge and I'm sure that you're all sitting on the edge of your seats because you also want to participate in this discussion. You can and absolutely should do that. Please use the chat on the platform where you're watching the show, let that be YouTube, LinkedIn, whatever else that you found us on, and just throw your comments, questions into the chat and we will make sure to try to reflect and answer as many of them during the live show as we can. And now we can go back to this whole edge computing thing. This is a very broad area and topic. So in that sense, it can be a little intimidating. If you just think about that we can focus on one edge, multiple edges, we can talk about the underlying infrastructure that is most often a mess of the distributed system. We can talk about management challenges and all these kind of things. We can also even just argue about is it really latency that is the main driver and requirement of Edge? I don't know. We will discuss things like that today. One thing though is sure there are more and more connected edge devices every day. In fact, according to a GSMA study, we are expecting 24 billion edge connections by 2025. That is a huge number. So you can imagine if you just try to visualize this number that how much demand it puts on the underlying infrastructure and how much work that is to really ensure that those connections remain live, managed, and can be operated in an automated fashion. So it is a challenging space. It is also a really exciting space in terms of how many new opportunities these connections will open up when it comes to applications and use cases. So how do we get to this 24 billion edge connections? This is where Intelligent Edge is coming into the picture that we will be talking about today. VinDriver concluded the survey with the support from the opening for foundation to find out why people and organizations are tuning into and using Intelligent Edge, what kind of challenges they are facing when it comes to execution and solutions. So today we will give you some of the highlights of the survey. We did have a previous webinar that goes through more of the details. So you can check that out after the show as well. Today we will give you some of the highlights and we'll also talk a bit more about open source and the solution space. So kind of give you ideas about the execution part as well. So before we jump into the numbers, I would like to turn to Paul first just to explain us a little bit what Intelligent Edge is exactly. Why is it in the spotlight? Why you felt like we have to run this survey? So please just give us a little bit of background before deep diving into the detail. Yeah, thanks, Odeko. So it's really, really great to be here today. Really excited for what we're going to talk about. This is a really important topic to us as a company and as an open source contributor. The Intelligent Edge we see, as you mentioned, the 24 billion devices that will be in place by 2025, those are connections at the edge. And they're really coming from the real new use cases that are being enabled at Edge. If you think about things like autonomous drone delivery and autonomous vehicles, we kind of all expect to have self-driving cars at some point. Even within that domain, there's the capability for vehicle-to-vehicle accident avoidance. And if you think about those types of things that are using machine learning and AI at the Edge where two vehicles need an ultra low latency connection with each other in order to avoid each other from an accident perspective, right? You think about telemedicine and the ability to run remote operating robots across the globe from a surgeon on the other side of the world. This is some really exciting growth. And a lot of these things that we all expect to experience as consumers are dependent upon the Edge. And even within that statistic that we had those 24 billion devices, 20% of those are going to be connected by 5G in that same timeframe in 2025. So this kind of paradigm change, if you will, where a lot of the applications we want to have is driving compute to the edge of the network, right, and out to the device. And it's driving intelligence with respect to machine learning AI and augmented reality and compute processing of functions towards the Edge of the network in order to build these applications. And that's fundamentally what the Intelligent Edge is about. Before going into the survey results, can you also tell us where did you run the survey just to give the audience an idea in terms of the responses like which areas, both in terms of industries as well as geographical locations? Sure. So we commissioned this third-party study in partnership with the Open Infrastructure Foundation. And we had a total of 578 responses, so pretty large survey. We had over 200 responses in the United States. We had 378 responses from France, Germany, Spain, UK. And we surveyed a variety of industries. So we looked into aerospace and defense, automotive, both the OEM and supply chain aspects, energy production and industrial manufacturing, medical and general technology that would cover manufacturing and hardware related things as well as telecommunications. And the reason for the breadth of that was really to look at, you know, as we talked about some of these use cases, you really have to touch all these industries to discover what's happening with these Edge applications. And so that's why we went so wide with that survey that it produced a lot of interesting information. I love this personally because, well, I've been involved in Edge for a little while now, and we have been talking a lot about it from telecommunications perspective, because obviously they are the ones who are having the backbone networks to be able to build on and creating these connections. But I think it is amazing that we are approaching the time when we can talk about manufacturing, when we can talk about healthcare, we can talk about all these other industries that will be enabled by Edge. So I think this is now time to jump into the survey results. I know that we already have some incoming questions. So I think as we are approaching some of the survey highlights, we can touch base on those too. So I would just ask first that how did the survey results reflect the views that you just shared earlier and how important people and respondents found in intelligent Edge is? I think it was really interesting. It kind of brought to the forefront some particular use cases, right? I think looking at the data from the survey, we had a variety of different use cases that the respondents spoke to, and at least 40% of the European and US respondents indicated that some investments would be actually increased by over 50% in certain areas. And there were six top items that you can see. If you watched that video that you recommended at the beginning, those were autonomous and collaborative robotics, artificial intelligence, really a hot area for the industry, additive manufacturing and 3D printing kind of surfaced as an Edge application. A really important one that we see, remote operations and maintenance, right? If you think about the modernization of the energy grid, which is going from a centralized model to one that uses a lot of wind and solar as a distributed model, that's, you know, requires a lot of remote operations and maintenance with using virtualization technologies. Another important one, artificial intelligence and machine learning for visual processing. You think security systems and that sort of thing. And then just the general topic of machine learning. And those really interesting results from the survey, because as you hear those applications, they make sense, right? These are hot technology areas. There's a lot of open source innovation happening here. And it's driving a lot of the applications that we talked about. Before moving forward, I just want to make sure that if Mark has some thoughts to any of these parts, he can chime in. Sure. Yes. Well, I mean, I think Paul's hit the nail on the head. I mean, when we talk about the actual applications, I think that that's what's really interesting. You know, we all probably dream about autonomous cars and all these other types of autonomous vehicles, like the Jetsons is finally getting closer to reality. And, you know, I think it's, to me, that's much more engaging and exciting than just thinking about the technologies. It's like, okay, it's edge computing. It's a category. There's, you know, analysts that study it and talk about it. Like, what's the real world impact in this whole, like, space of industrial companies and applications and the factories? Like, it's really starting to become the hottest area when we talk about, like, open source adoption, things that we would have thought of originally as sort of cloud computing technologies. But they're not, you know, up in the cloud, so to speak, it's being pushed to the edge. And that's really awesome, because we're seeing a lot of sort of reuse of technologies that were proven and built in sort of the centralized data center era. And really all those were, if the fundamental level is compute storage and networking, how do you automate that? How do you manage its scale? But now we're taking thousands of sites to billions of connections. It completely changes, you know, the difficulty and are the bottlenecks and a lot of work has to go on. You can't just drop, you know, the, you know, open stack from five years ago into a factory and have it control robotics, right? Like, so there's so much we can reuse about it, but so much more work that's gone on and will go on for, you know, we talked about sort of the next decade of open infrastructure as this incredible ground for innovation and collaboration across the world and companies and countries. So that's just an exciting kind of next phase when we think about, you know, where we're going as a community. Yeah, you touched on some really important things there, Mark. And I think that some of the things you mentioned, they're actually the genesis, the reason for Starling X within the Open Infrastructure Foundation, you know, why create a completely new open source initiative? And there's some real reasons behind that. As you walk through that, you think about these thousands of distributed sites. You think about a service provider deploying virtual 5G infrastructure or, you know, as you mentioned in manufacturing with robotics and lights out manufacturing or augmented reality for servicing things, you're talking thousands of distributed sites. And we, of course, are coming off of a bunch of years of activity around open stack and use of Kubernetes in those environments, right, as we move towards more cloud native approaches. However, they were really, we built those technologies together as a monolithic environment, right? It was running one data center. And we had a lot of challenges getting, you know, Nova Glance, Neutron, Keystone, all these components to work together and build the virtualization tool set that we would use to build the private cloud, right? But now as we look towards the new problem of a highly distributed system, it's a distributed cloud with edge compute. Now, in order to run that, if you're a business trying to run that, you've got thousands of sites you have to manage as one distributed environment. And that fundamental new set of requirements is what drove the existence of Starling X to say, how do we reuse technologies like open stack and Kubernetes and solve that distributed problem? And new innovative ideas happened in that open source activity around distributed control planes and self healing networks and things like that, that fully embraced the problem of a distributed edge cloud. And that, that to me is what's exciting about Starling X is it's kind of the next generation that still has its foundation, open stack and Kubernetes and those type of technologies, but is now taking a fresh look at the way you deploy highly distributed systems. And we will be talking about Starling X a little bit later. Before we go into further details about the project, I would go back a bit to Intelligent Edge. We talked about the application areas. I wonder what the survey results said about benefits of using Intelligent Edge. Yeah. So, Mark, I don't know if you're going to take any of these, but we have a lot of, a lot of really interesting information. I happen, you know, I don't know this all by heart. I'm looking at, you know, my screen here with some notes that we have from the survey because it's really interesting data and there's a lot of it. And I, I certainly encourage everyone to look at the Open Infrastructure Foundation video that we did together that kind of goes through this in detail if you're interested in it. But really interesting things, you know, 74% of the respondents, right? Fully three quarters of them are looking to get more insights to improve their operations. The stat on the screen here, 76% of the respondents expect this to be where the opportunities for innovation are going to surface, right? And I think that as we talked about with Telemedicine and Autonomous Vehicles and all this sort of thing, that's why, right? That's where the innovation's happening. It's no longer in the central data center. It's at the edge of the network. We think that, you know, the data showed that as we all talk about digital transformation in our companies coming out of COVID with this, you know, the tools that we're using right here to communicate with people are very different than, you know, how we would tend to physically gather previously before. And 72% of the respondents expect the rate of digital transformation to be accelerated by the adoption of edge compute. So, boy, the intelligent edge and what's happening here, even the reduction of OPEX, right? You look at these technologies we're all working together in the open source community, they're heavily dominated by artificial intelligence and machine learning and high levels of automation. And that drives OPEX down for businesses. So, a lot of great statistics to kind of cement why this is happening at the intelligent edge and what the benefits of it are from a response from that survey. Yeah, and I think that, you know, why I think it's so interesting that in this particular case, you know, people, the top reason that they're looking at intelligent edge and they're investing in it, some of them are going to increase their investment by 50%, is because this isn't just a case of some sort of solve problems that they've had maybe some expensive solution to, and now they're like, well, how do we like just kind of reimplement it with open source and take this thing and sort of commoditize it? This isn't really that at all. This isn't like, okay, we're going to take, you know, office and turn it into Libre office or something, like that's sort of the open source of 10 years ago, right? This is about innovation, it's about where companies are expecting to have differentiation or really generate value for their customers. It's where they're trying to do things that have never been done before. And that's why they're willing to make an investment in it, right? Like you said, I think some of the people said they were going to increase their investment 50%. They're hiring people, they're looking for solutions in the market, they're looking to influence those solutions. They've got certain challenges ahead of them because, and the reason people are willing to tackle these challenges because they're expecting it to be really the cornerstone of how these companies deliver value to the market, right? And so we see there's that the conversation from several years ago about how software is eating the world and people say, you know, every company is a technology company now today. I think this absolutely reflects that trend, this data just really underscores that, that we're seeing and believing, but it's coming through and the companies you're talking to. And you talk, when you go through that list of industries that you mentioned earlier, right, it's like manufacturing. I mean, these are industries that have been around a long time and they're having to completely change the way that they do what they do. You know, lights out manufacturing. I mean, it's a completely different world than manufacturing from a few years ago. So you can think of those as kind of like, you know, Old Guard, big legacy companies, right? And those companies need this type of innovation, this technology, kind of transformation top to bottom and pushing it to the edge is exactly, you know, why we're excited to work on all this stuff. And it is, it's a big enough problem that affects so many different companies out there that we need to work together on it. Is it, you know, I'm going to keep coming back to the open source part of it, because I guess I'm a little biased, open in for a foundation and all that. But it really does work when you have problems with this magnitude, like no companies can be able to solve all this on their own. Yeah, I completely agree, Mark. I think, you know, we're really excited being an active member in the Open Infrastructure Foundation with what we really view as a place for crowdsourced innovation, right? The ability for multiple companies and individuals, developers to come together and contribute in a way that can solve these problems at scale, right? It's, if you look at the legacy of OpenStack, it's incredible, right? Look at the years and years of development and the, you know, 100,000 developers that poured energy into that, that technology. This is what we're trying to do here for Edge, right? We know that it's, if anything, even a bigger problem and a bigger challenge from a technology stack perspective. So the only way to solve it is really via open sources. The only way we'll get the scale and the innovation that we need to solve these problems. And I think where the discussion is going right now, it's showing really well that it's not just all sunshine and rainbows. We do have challenges when it comes to implementation. Before going into the survey results part, we do have some questions in the chat already. So I thought to throw in the, like, what is your view of security and privacy at the Edge? I think it's good if we touch on this one here and we can always come back to it when we gather our thoughts on the solutions part in the second half. Yeah, I think you're right in that it's not all roses and happiness, right? Security is a significant challenge. And obviously when you go to a highly distributed system, it increases the threat surface of that system. It gives more points where people can attack that system. And it continues to be an area of active investment, both within the open source community and within private community, in order to ensure that the architecture is hardened to those type of challenges. And some of the things that we've done around central of declarative provisioning and the ability to audit and maintain Edge sites with the synchronized control plan, they contribute toward the ability, also with the secure and crypto and key handling that's being done throughout the system. So definitely an area of focus and something very important and a challenge for everyone going forward. To add to that, what other challenges did the survey highlight? Would you mention any of those? I think that I'm going from memory here, but certainly from what I've also encountered and directly talking with people that try to use this technology is there's one set of challenges about building the technology stack itself and getting it to function correctly. But the real challenge is operational, right? And I think I mentioned earlier that 72% expectation to reduce OPEX, that's coming from automation. And I think if you don't embrace automation, orchestration, things like zero touch deployment, lights out management, et cetera, of remote systems, think about a company trying to deploy these thousands of sites. They've got to physically deploy them and maintain them. And it doesn't scale having somebody at a terminal deploying the sites individually. You've got to have a high level of automation in the solution. So I think that keeping an eye on OPEX and thinking about the operational tools, the ease, the burden to run and deploy the system is a real important piece of this. And I think Ildeco, another thing that I know came up in the survey is, and it's interesting because I think I hear this from every company I talk to these days, kind of across industries really. And when it comes to technology and transformation of their businesses is the skills gap and knowledge gap, right? I think it was the number one cited barrier was when we asked people, what is it that you are concerned about that's going to keep you from being able to embrace the intelligent edge? And it was like 34% or looking for more skills, more knowledge sharing. And I think this hunger to hire more people, train up. This is that challenge, right? As companies just trying to complete transform, become technology companies, become, you know, take that software that's eating the world and figure out what to do with it, operationalize it. Like as Paul said, it's not just like is that does the technology exist? Can I buy it? But can I operate it? And I think that this was the number one cited barrier or concern. And I think it's not just people with intelligent edge. Everybody I talked to out there is really trying to get more knowledge, more skills, hiring, you can see companies throwing money around trying to poach people. Like it's just a hot job market right now because this is kind of the thing that you need, right? And part of the automation is also tackling that. And so I think that that's also shows up in the data in my opinion, correlation here with open source. And one of the things that was asked in the survey was, you know, what are you looking for in terms of solutions for intelligent edge? And what the data showed was that people were preferring open source solutions over proprietary solutions two to one. So that was a pretty huge kind of landslide victory for open source over proprietary just going across. And as Paul mentioned, again, like all those industries, you know, they're not necessarily the industries that historically you think of as leading the charge and technology or leading the charge and open source or really, you know, getting out there over their skis and trying to do these really challenging tasks in terms of transforming their companies with these technologies. But it's no contest these days. You know, I think a few years ago this would have probably been the other way around. But I think that the reason for that in my opinion, is it does come down to that access to knowledge, access to talent. And there's a realization that no company in the world can possibly have all the smartest engineers, right? And if you can tap into a community and be part of a community in which you your engineers can talk to engineers from other companies, perhaps they're in other countries, right? Like the world has so many smart people all over the place working for a lot of different companies. And we can advance the whole industry, the whole planet really in terms of how we make progress, get on my soapbox a little bit here. If we do work across companies, and that this is this is I think why companies are embracing open source as a strategy. And it isn't something you just do overnight, right? I think Paul can speak to this. He's been doing an amazing job of leading wind river into a company that's open source first. And it doesn't happen overnight, but they've been, big leader in Starling X and embracing open set Kubernetes. So I think that it is those two things are related, those stats. I don't know, Paul, what do you think about my theory here? No, I think you're dead on. And there's a couple subtle things there as well too that support the statements you made. One is that the way that you do open source is also very important, right? And one of the things we've adopted as a company, and it's hard for a company make this transition kind of mentally to, is this a safe thing to do for our business and all that kind of thing? But we are what we call an upstream first methodology open source company. And what that means is every bug that we fix every feature that we write goes directly into the open source first, right? And that I think is the right kind of community cultural approach for businesses to take. And I know that can be scary because a lot of businesses might say, well, I have this idea. I want to protect it. It's my innovation. But the reality is the growth and success of your business is from that community open contribution and crowdsourcing, right? And so that's one thing is you have to have the right approach to open source, right? And truly embrace it, right? And that's a hard thing for business in particular. It's easy for individuals. It's a hard thing for business to do, right? The other thing that, you know, in support of your statements around the one business can't solve it by itself. And it's kind of a necessity because of the scale of these problems to drive this into a community activity. There's also a business reason from the customers. The customers are starting to get to the point that there's commercial reasons why they want this, right? That, you know, if Company A starts disappointing them, they can switch to Company B and still get the same technology, right? And so there's this, it's healthy competitively to do that, right? And it enables them in particular telecommunications industry. We've seen that, you know, there used to be, if, you know, we talked about 5G a little bit earlier, in 2G, 3G, and 4G, there's a lot of competition at the initial win of the network from all the vendors. But then that's it. There's a vendor that's in there and the deploying physical hardware and that's it. Now that you go to a complete virtualized open source approach for the life of the network, these are now software applications, right? And a vendor can be swapped in and out at a whim of the carrier. And this gives them a tremendous amount of commercial power and it's only happening because of open source. So I think you're seeing not only because of the scale of the problems like you articulated, Mark, but the business reasons from the customers. This is the way this is going to be done going forward. Yeah. And I think there was a time when people perhaps even, you know, rationally thought of open source as the riskier path. And now I think it's a much bigger risk as a company, as a customer who's deploying this stuff. If you don't embrace open source for all the reasons that you said, there are hardcore business reasons, right? It's not just, hey, we love, you know, lines of Python and we want to get down in there and change the code. You know, even if you have no interest in that whatsoever, you're not planning to kind of, you know, as an end user who's adopting it, kind of modify the code and do all that stuff. There's tons of reasons just purely from a business practical standpoint to run your business. You know, intelligently, open source is a better business strategy in top of just like we all think of it as a better way to solve these problems. Like you said, they're more complex than a company, anyone company can tackle to solve. But that's certainly true of just anyone vendor trying to solve it, right? But as the end user, even if you're just thinking about it from, you know, your CFO standpoint or your procurement team standpoint or these kind of sourcing of the, sourcing of the technology that's in your supply chain, like it's actually the dominant path these days for good business reasons as well. Yeah, I think, and just as a quick follow up, you know, a company like ourselves, I know what we live every day and it's really our offer to our customer centers more around our expertise. We know how the network works. We can help them build things that we can provide support and maintenance. We have the scale and experience to provide a version of that open source that can be backed up by a commercial entity and the customers still want that, right? In other words, the reason for saying that is diving deep into open source and making that commitment to contribute there doesn't break your ability as a business to generate revenue from that open source, right? And that's a really important message that, you know, don't be afraid to jump in and participate because your business and the expertise you build around that open source gives you the ability to grow revenue, right, as a contributor. Yeah, that's a good point. I know we're just kind of going off on this ill to go, keep us back on track here. But it's just such, it's a cool topic near and dear to our hearts. I think, you know, that's another sort of myth maybe and hopefully it's dying away, but that like somehow open source is incompatible with, you know, making money or capitalism or whatever. I mean, I'm an unabashed capitalist, I'll probably get some hate mail for that. But, you know, I also love open source and because they actually go really well together, contrary to what maybe people thought a few years ago with the kind of stereotypes around open source. It's not an either or by any means, it's generating a lot of value and companies are rightly getting paid for that. And so it's not a choice between those two. That's a false choice. Yeah, so here I would like to really just echo what the two of you said in terms of how important it is when it comes to open source to participate in these communities and not just picking up an open source software and the code, but being part of the community, share your views, feedback, helps steer the direction of the project. And when it comes to participation, you can also participate in this opening for live episode. So don't forget to throw your comments and questions into the chat. And here before we go into more the technical side of solutions and the project themselves, I would like to pick up some of the questions from the chat that we got earlier, because I think they are interesting and are pointing towards the solution and implementation part of our conversation. So one of these would be, do you envision a private edge infrastructure or do you think enterprises will start their edge journey hybrid by default? And by this they mean whether it's always with a tackle operator and or hyperscaler partnership? Yeah, I guess I'll start, Mark, that it's a heck of a complex story there where it's very industry dependent. If you think about telecommunications industry, at the far edge of the network they own the physical footprint, cell towers and the remote sites and everything. And of course the hyperscalers, the public cloud providers, are coming into what they consider the edge, which is really the core of the carrier network with mech solutions, content delivery, gaming solutions. There's a place where the hyperscaler can come in there and also we're seeing some carriers move some of their infrastructure into the public cloud. But in the end as you progress towards the edge of the network, this intelligent edge, and then jump to the device, that's really a privately owned infrastructure today and I think will be for quite some time because of a lot of technology and business reasons. As well as you think, I gave an example earlier about the energy sector, in particular in France, there's a big initiative now to go to distributed power grid with solar and wind power. Again, a power infrastructure utility like that is going to be government and privately controlled. It's not going to be a public provider that comes in there, but at the same time the OSS, BSS and billing systems and top level management could be in a public cloud. I view the hyperscalers as synergistic with what we're trying to do. It's about a core to edge architecture and they're more dominant at the core and private, high performance, low latency, real time is more dominant at the edge and the two things kind of flow into each other as those partnerships evolve. Of course we'll see that change over time as the different businesses try to succeed with certain initiatives, but in general I think that's how we look at that. Yeah, I mean I don't have a lot to add other than just, I mean it's a great question, but I think what I've found throughout my career is that there's often a desire to kind of answer this question of is this new model at the expense of the old models, this new technology at the expense of the old technology and the answer is always never that simple and usually just no. It's usually nothing ever seems to replace anything. It's the sort of layers and additive and things that we think are going to suddenly displace that the previous thing that was getting adoption in terms of the model, the technology of the company, oftentimes it just becomes dependent on the industry, dependent on the customer. I wish that there was just a simple answer, but I think as Paul said people are really deploying these edge technologies to solve the specific problems, but in many cases they're by no means going to stop relying heavily on the central sort of clouds, whether they're hyperscalers or whatever, so it's not maybe like the snappiest answer because it's not, it's just complex, but it's a good question. I think Paul hit the nail on the head. Raul also has another question that kind of just adds to the convergence between edge from operators and edge from hyperscalers. I know that Paul you touched on this a little and the synergies, so before we move forward I just wanted to ask if you have anything more to add on that. Yeah I guess the key thing that we're seeing is obviously the the statistics we talked about in this survey are now becoming, that knowledge is more widespread, right? The people in the industry are understanding the opportunity. The survey that we did in communicating it here kind of broadens the people that know about this, but a lot of people in the industry are very aware that edge compute and intelligent edge systems of where growth is, right? And so I think that hyperscalers are no different and they're trying to push towards the edge you may have seen with the Amazon wavelength and outpost or Azure, you have edge zones and arc, right? There's different technologies that are being used to move towards the edge of the network to embrace that growth and that opportunity. At the same time as you get to what we consider to be the far edge of infrastructure or the device edge where you've got robots and automobiles and drones as the device is connecting in as well as that far edge infrastructure, that's not yet a place where hyperscalers play. So I think again it's a core to edge topology. It's an end-to-end network and each of these players has their role to play in that solution. Awesome. I think this kind of the discussion ties into jumping over to the solutions and projects from the survey as we kind of started to talk about private edge, not private edge, what infrastructure you're using there. And Paul, you already talked a bit about Starling X. So I think we could jump over to talk a bit more about that project. For those of you who do not know Starling X yet, it is an open-source cloud platform that is high-tuned and optimized for edge and IoT use cases. This platform is also creating a fusion between open stack and Kubernetes. You can run all kinds of workloads on it. I was also mentioning things like how you're managing a distributed edge infrastructure. Starling X also has a really cool feature called distributed cloud that kind of gives autonomy to your edge sites but still synchronizing the relevant data between the core and the edge as well as you can also choose to run only containerized workloads on an edge site really kind of fine-tune the system for your needs. It is putting a high focus on areas such as security. I know that there is a question about security in open infrastructure so we can deep dive into that as well in connection to this. But before going there, I wanted to mention that we had an open-infra-light episode, another one around edge, where we did talk about the latest Starling X release, the 5.0. So you can find that in our playlist on YouTube and go back to learn a bit more about the technical details and the cool new features. But as we have Paul with us today and Windriver is very involved in the project, I just wanted to ask you, Paul, if you could talk a bit more about your involvement why Starling X is important for you? I'm glad to do that. I mean we have, we're a business that's been in business for 40 years, pure software company the entire time and our history is in that of embedded devices, real-time systems, we do avionics, automobiles, robots, right? Those are the things that are the devices that are now being connected to the intelligent edge. So as the intelligent edge and the systems around this kind of exploded in recent years, we found ourselves in the spotlight of a real growth in all these new use cases, right? The autonomous vehicles, telemedicine, all of these things. And similarly, we've made the investment both in OpenStack and the OpenStack community as well as more recently in Starling X, that was to really embrace the opportunity around distributed cloud and edge compute. And suddenly we find these worlds crashing together, right? And for our company, it's a real great opportunity space to bring the device edge and the infrastructure edge together into what we call the intelligent edge. And so as a result, it's very, very important to our business. And as we've talked about with these survey results, if you see 76% of respondents recognizing this is where the applications and innovation are going, a company like ours is going to follow that direction for business opportunity. And so the investments we've made in open source, both with Linux, with OpenStack, with Starling X are key to our business success growing forward around the intelligent edge. And just in connection to all this, I just saw a comment from Scott on the chat that they are implementing Starling X this week and deploying it. I think it is really exciting. So in connection to... Users are coming in as we speak. This is great. Yes, it is amazing. So I wanted to further ask the question if you can share some information about users of Starling X. Yeah, I think we're kind of reaching a really exciting time in Starling X in that the project just graduated last year. It's actually very new. There's a lot of innovation and excitement happening here. It's reached a level of maturity that is now in production deployment with several customers. Folks may have seen the press release that was related to Wind River. We participated with other vendors like Intel and Samsung in the world's first 5G fully virtualized call. And this is using virtualized RAN technology at the edge of the network. This is the first time that in telecommunications all the way to the edge of the network where the cell towers exist. It's now fully virtualized and in fact Kubernetes and cloud-native-based infrastructure. And so that was a use of Starling X. And we continue to grow that deployment with Verizon. And we're very excited as a company to earn their business. And we see great success happening from the hardening of Starling X. And as I mentioned, the open source approach we take where everything that happens that hardens that in that deployment, we pour back into the open source community, right? And others get to benefit from that improvement in stability and quality that happens in the software. Recently also, a lot of activity kind of in a parallel environment is the open source activities around O-RAN and the O-RAN SC and software community there. Starling X has become the foundation of what's called O-Cloud in that open source community. And that means it's the reference for fully virtualized RAN deployment now globally for open source. And you may have also seen the Vodafone announcement where Vodafone as a result of all that and their active interest in open RAN as a kind of way to get all the vendors to interact with each other from an API standpoint in the RAN technology has now selected our technology based on Starling X for that deployment, right? So another customer there going to deployment. We've had a really exciting customer that we've worked with also public T systems and they're doing with their Edge Air offering for Edge Compute for commercial enterprises think manufacturing augmented reality and really exciting things happening there in the commercial sector around enterprise use cases for Edge Compute. So that's the stage we're in now, right? The project's graduated. It's stable enough that customers are starting to deploy it and it's a really great time to jump in and participate in that exciting growth that happens from that, you know, that kind of really nice inflection point that an open source project has when it becomes kind of real, I guess is the right thing. You know, in the early days it's like, hey, this is a great idea. Let's go play with it. But now it's a real thing. It's in production deployment with customers using it. So you're talking about Verizon, Vodafone, T systems. I mean, these are companies that I think, you know, probably everybody's heard of and they're different parts of the world. But if you haven't, you know, these are massive, massive players in telecom and you're talking about Verizon and I think some of the other examples specifically about the RAN. And I think this is really cool to me. The early days of OpenStack, first of all, we had no clue that it would ever get used by any telecom operators for anything. And the fact that like, like nine out of 10, nine of the top 10 telcos in the world, you know, run OpenStack and other, and we think of that more sort of in terms of the traditional compute and that it's moving in to the network and that moving the packets for the network. But the RAN is a whole nother level that again, we have no idea that it would eventually kind of be utilized in that way. But maybe you can explain what is a RAN? Because to me, I mean, that's something I had to learn about over the last couple of years as Starling X came along, when River started getting these awesome wins, you know, Verizon's RAN for their 5G, like what is a RAN? Yeah, so RAN stands for Radio Access Network. And in a 5G context, that's broken down to a set of logical software functions, the RU, CU and DU in particular, as you move towards the edge of the network, it starts at the CU right after the 5G core. 5G core, of course, has an architecture that the 3GPP has standardized. You can go look at all the software components in that architecture at that standards community. It then goes, flows down to the CU centralized unit, the DU, the distributed unit, the DU is directly driving the radio unit at the cell tower. So for the first time ever, you've now got service providers moving to an architecture where every single asset is becoming software-based, that they use to run their networks from the core. And there's a lot, as you mentioned, a ton of OpenStack use there as that industry went through a transformation called NFV or network functions virtualization years ago. And so OpenStack and virtual machine usage is dominant as they transform the core of their network. Now RAN or VRAN, the virtualized RAN, which is the way of doing it with pure software functions, is bringing virtualization both with OpenStack and with cloud native approaches like container de-run time and Kubernetes orchestrators, et cetera, out to the very far edge of the network. And what StarlingX is doing for the first time is combining both of those technologies and embracing the distributed problem. StarlingX is a little unique in that it supports a containerized version of OpenStack inside of Kubernetes framework. And this allows people to run core data centers with high VM scale because the VMs aren't going away anytime soon. There's been a lot of investment to move things to virtual machines, so you still need that high-scale VM support. But as you go to the edge of the network, people are building things with a more contemporary approach of cloud native microservices, containerized architectures. And so Kubernetes and container de-run times, for example, tend to be more utilized at the edge of the network where green field activity is happening. And so that's what RAN is and how VRAN kind of maps into the technologies we're talking about from the Open Infrastructure Foundation. Yeah. And so no RAN, no cell phone. That's a good way to think about it. But I think, you know, and also what you are just describing, I think, is a really good illustration of something I mentioned earlier, which is no new technology just like replaces the old one. Like, it's what we think it's going to do, but it never does. Like, did Kubernetes replace OpenStack? Absolutely not. People are combining them. Does StarlingX replace OpenStack and Kubernetes? No. StarlingX incorporates the OpenStack and Kubernetes puts them together for specific use case. Now it's going all the way to the radio access network level and VRAN. So did containers replace virtual machines? Absolutely not. They're being combined together every single day. And sometimes that overly simplistic narrative that people like to put on Twitter to get clicks or whatever, it never plays out that way in the real world. And I think it's so awesome to have you on because you've got, you're building real solutions, you have real customers, you've got Verizon, T systems, Vodafone. And these are people. And then we, as you were talking earlier, kind of the future looking forward to manufacturing and all these autonomous things. Like that's how technology actually gets adopted in the real world. It's not the way that people think about it on Twitter as much as I love Twitter. You know, it's a little reductionist and sometimes overly kind of pitting different communities or technologies against each other. And just, it just isn't how it works in the real world. And so I love hearing about these real deployments. And so anyway, it gets me excited. I think you're right. We tend to look at, oh, this is the next wave of technology, the previous wave is dead, right? It is not like that, right? There's kind of these gradual transitions and overlaps that happen over time. And you get to use, you know, if you look at OpenStack in particular, how many years of deployment, innovation, and hardening has that technology been through as an open source activity? It is an exceptionally robust technology used by thousands of businesses today. Why not leverage that, right? It's an important foundational element of what we need going forward. It isn't replaced by things like Kubernetes. It's augmented and compatible with, right? And so we look at it that way. As you move through this kind of open source ecosystem and landscape, you get to pull from all of these different projects and innovative communities and build a solution that solves the real problem, you know, to enable those new applications. And actually, Terry Perez has an awesome post about this, about technology being additive. If you go find his blog, we can send it on Twitter later if you want to get on Twitter. But anyway, it's a cool topic, I think. And I think on this train of thought, we mentioned a couple of open source projects and just kind of talking about how they integrate and how they innovate together. I think it is a good time to kind of talk a bit more about these adjacent communities as well, and just kind of showing the landscape. So I would drop the ball over to Mark to talk a bit about the opening for projects. Yeah, yeah, sure. So, yeah, so open in for foundation, you know, our mission is to, you know, we build communities who write software that runs in production, specifically for infrastructure, obviously, and that started off life as, you know, OpenStack being the first project that we were helping to get off the ground. And now, as Paul said, it's, you know, had, I think, 23 on-time releases now. We're almost 11 years. Actually, we're a few days away from the 11th birthday. So that's obviously the heart of open infrastructure from our work as a community and as a foundation, supporting that community, building software, running production. But there are a lot of other projects as well. If you go to openinfo.dev, you can actually learn about all these projects. Or if you want to join the foundation, you can go to openinfo.dev slash join, got the URL up there. But, you know, each of these projects is complementary in many ways. And as we've been talking about Starling X, which is here on the slide, you know, as mentioned, it brings in both OpenStack and Kubernetes and things like Ceph and other technologies. You know, we have other projects like Zool, which is for CICD. This actually came out of the OpenStack community because we were testing every single day just incredible amounts of code coming in. You know, OpenStack is one of the, even today after 23 releases, one of the three most active open source projects in the world as measured by merging code, amount of code being merged every day. And so to deal with that massive influx of code and to kind of help test it in a proactive way, we have the Zool community has their slogan is stop merging broken code. And the idea there is that all the different commits before they even merge, get tested and have to pass a set of tests. And that's a really good way to raise the quality of your code. And now we see companies like BMW, Volvo, GoDaddy, all around the world, Le Boncoin, which is one of the largest e-commerce companies in France adopting Zool. So it's no longer just for the OpenStack community. It's been put out there and lots of people are adopting it. I can't go into detail on all these in the last couple of minutes, but you know, as I said, openinfo.dev will link out to all the project websites. And you can learn about them. Cota Containers is a good example of how both containers and VMs are sort of not, one's not replacing the other. Cota Containers is a secure container system that gives you kind of the speed of containers, but it also has the isolation elements of virtual machines. Airships, another awesome technology out there that's being used by AT&T and many others, SK Telecom to automate their deployments for their networks. And 4G and 5G and OpenInfo Labs is this awesome collaboration of universities, including Harvard and MIT and many other awesome universities that you've heard of, really testing these different types of technologies in production, a bunch of the universities around the US. Those students every day log in and run their workloads from all the different departments of these universities on production clouds and the OpenInfo Labs team actually is able to get real-world experience and feed that back into the upstream community. So the idea here is we're not just sort of testing the code in the abstract upstream, but also in a production environment. A lot of telemetry comes out of that, which improves hardware and software that's used. So that's kind of just a very, very quick overview of some of these projects. But as I said that, you know, as a community, as a foundation, our mission is to build communities who write software that runs their production. So if you have a project in the space, you know, we'd be happy to help you with it, maybe a good fit. The foundation can support your community building efforts and help you build software that might be running the next network for 6G. Who knows what's next? That's kind of a very quick rundown of some of the projects that we host within this foundation. I would also just tie it in here as it comes to integration and collaboration that, for instance, starting X is supporting Cata containers as a container runtime and you can use it with time-sensitive networking support. It also integrates components from Airship to deliver the containerized OpenStack services and the deployment of containerized infrastructure and starting X is also using Zool just to, yes, we are not merging broken code. So I think just on this slide is an amazing amount of integration and collaboration going on besides what we mentioned already earlier during this episode. We are sort of running out of time soon. So I just wanted to ask Paul if because the starting X community is also collaborating with others in the broader open source ecosystem, if you want to give some more examples to that in like two minutes. Yeah, I guess I can do that quickly. Of course, in the same way that we consume components from Airship like Armada, we use OpenStack within Starling X itself, right? We take Kubernetes and CNCF components like Calico downstream into the projects, a lot of interaction between open source communities within Starling X and the OIF. It's also, as I mentioned much earlier, excitingly for us in the telco domain, the foundation of ORAN SC for the Ocloud infrastructure for the different what are called VRAN split models. These components I mentioned in VRAN have different ways of being virtualized and interact. So that's a very important component there and also being used within TIP, which is the Telcom Infra project as well. So people are starting to see the kind of attraction of embracing that distributed cloud problem and using it for those kind of applications in other open source communities as well. So its use is growing in those communities. Amazing. Thank you. We have like two minutes left. I see one question from Scott that I can give a very short answer to. Namely, will R6, like the 6.0 release, move away from CentOS? There are efforts that are exploring other operating system options. For instance, Debian. So there is already work on going on in this area. I cannot, then the community cannot yet guarantee that it will fit into the 6.0 timeframe. So it will probably be a time box effort. But yes, the community is looking into both multi-operating system support as well as alternative to CentOS. So for further information and details, I would like to encourage Scott and everyone else to subscribe to the starting X mailing list, starting X-Discuss. Check out the website for more information and ways to get in touch with the community. And as always, a call to action. Participate, participate, participate because we are all better together. And the remaining roughly 51 seconds. I just want to give an opportunity to both Paul and Mark, if you have like a one or two sentences, conclusion to this episode or anything that you would like to highlight. Then, Florie, maybe start with Paul. I guess just super quick. We've talked about the intelligent edge, real exciting applications, telemedicine, drones, autonomous vehicles, manufacturing, telco 5G. If you want to be part of this, come contribute in Starling X, this is where the innovation is happening. And I think you'll really have a great time. Yeah, I mean, just exactly the same point. I would reflect that as well, which is that, yeah, if you're geeking out on all these incredible applications like I am, what's so awesome about it is literally anybody in the world can be a part of this. You can be a part of shaping the technology that run these crazy robots that are going to be flying around the world and self-driving cars that don't crash. If that just makes your imagination go wild, it's not just imagination, right? This is something you can contribute to. And lastly, I'll just say, if you want to hit me up, I'm Sparky Collier on Twitter, and I think we want to make sure that we know how to get ahold of Paul as well, maybe hit him up on LinkedIn or wherever he wants to be found for follow-up. Yeah, thanks. And I think you'll also see on LinkedIn, just as a final poke at this, we have a ton of positions open. There's so much growth in this technology. If you want to work on Starling X, you can actually get paid for it. So if you're a software developer, take a look at our open positions we're hiring in this space. Nothing wrong with getting paid. Capitalism is okay. No, you gotta pay for the coffee, right? I mean, you gotta have coffee while you code, right? That's true. I love getting paid for what I love doing. So I'm hoping that many of our attendees will check out the open positions and also just get inspired and excited about getting involved in Starling X and other opening projects as well. And with that, we are already a little bit of overtime. So I would like to thank Paul and Mark are two amazing industry experts who joined today for this great episode. Thank you for being with us today and sharing all your knowledge and expertise and thoughts in this area. In here, as a wrap up, I would also like to mention that Windriver is also an opening for Foundation Platinum member. So a huge thank you to them as well as the rest of our member companies who made opening for life possible and make opening for life possible. If you're interested in joining the Foundation, you can check out openingfra.dapps.join. And last but not least, next week we have another amazing episode for you. Today we got only as far as touching just a bit on security and we didn't even get to arrive to the phase of data. We are all living online lives and much of our data is also living online. And with that data sovereignty is something that is more and more in the focus. So that is what industry experts on next week's episode will talk about. Namely, Johan Christensen, CEO of City Network, Pierre Grandlier, CEO of GaiaX, Kurt Garlove, CEO of Starving Cloud, Stack, and Linda Seif, Chief Commercial Officer at Binero. So if you are excited about this topic, tune in next week. As a quick reminder, it will be Thursday at 1400 UTC or if you would not be able to follow it live, then you can always check out the recording on our YouTube playlist. And one more call to action to you. If you have an exciting topic that you would like to share with the world, please throw your ideas into ideas.openinfra.live because we would love to hear from you as well, all of you around the globe. So mark your calendars for next week or ongoing for every Thursday. Thank you again, Paul and Mark, for being here and joining and thanks to all the attendees who followed us this morning, afternoon, evening, or night. And with that, see you all next week on Openinfra.live.