 Hello, everybody. Good afternoon. Good morning. Good evening. Welcome to the global audience. This session is on uniting the edge for tomorrow's demands. So I think thank you all for those who have joined. I'm going to cover essentially five things in the next 50 minutes. There is an overview, and then there's details. So please stick around. I'll keep a pace and allow you to take screenshots as well. Obviously, the most interesting part of the virtual events is screenshots without not taking a picture. So the topics I'll cover are the most pressing question, which is what is edge computing? Everybody has a different definition. What do we do? Is this an edge? Is that an edge? So we'll get to the bottom of that. Then the next big question comes around, what are the killer apps? What are the use cases? So we'll talk about that. And then, of course, we'll turn the majority of the presentation around what we are doing at the Linux Foundation for LF Edge, how we're growing, and then go deeper into each of the projects and then point you to the link on how to participate. So excited to be here. I know it's not in person, but I think we're now getting used to this new mode of learning. So let's get straight into it. What's the definition? You already know, since you're part of the OSS or open source community, that we are more than Linux as a Linux Foundation. I want to put it in context where Edge and IoT comes in. So we have several open source projects in several of these technology areas or several of these markets. And Edge and IoT is a big area of focus for the Linux Foundation. And we are hosting staffing projects that are extremely important in the Edge community. Some of you already have seen the cycle here where we host the project. The projects create developer community. You all participate. And then technology or software is created, which results in products into deployment. And then deployment leads to profit and the cycle repeats itself. So the reason I'm going to give you this background is in a couple of our projects or specifically a project called Acreno, we have a notion of what is called blueprints. And what blueprints do is they accelerate the bottom half of the cycle, right, from products to deployment. And then that allows for rapid profits and interoperability. So we are kind of doing testing and blueprints in open source. This has never been done historically, or if it was done, it was done kind of behind the closed door by end users. This is a new way of not just creating software and code, but getting code into production as fast as possible and as openly as possible. So keep that in mind because I'm going to bring the circle up again when we get to the project called Acreno. All right, let's start off with what's an Edge? Well, let's first understand this diagram. On the top left, you have an enterprise, you have a sensor, you have a micro controller, you have some version of an Edge, as we loosely call it, or a building. And that goes into some form of a macro connectivity, whether it's a base station or whatever. And then on the bottom left of the screen, you have the home, right? And we can all relate to that because we have connectivity going from our gateways back into a central office. And then you come into a telco and then there's a whole other universe that exists called cloud. And just note the 20 milliseconds here because that's a very important attribute that we have to sort of define. And so Edge has three parameters that we look very carefully. So the classic copy book definition of an Edge is, hey, I have proximity of compute and storage. So now that's an Edge. Well, even more important is the responsiveness attribute. And that's anything and in any application that requires roughly five to 20 milliseconds of latency is an Edge application. So I'll give an example. If I have a sensor that wakes up and dumps its analytics and data into a gateway every week, that's an IoT application, but that's not an Edge application. And then that's the responsiveness. And then of course, there's the mobility angle where connected cars and drones and all these new IoT or new connected devices that come up that are all interested in taking care of the Edge compute and storage closer to them. So that's keep these three parameters in mind because I'm going to double click on this in a lot more of these verticals. So each of these verticals, and this is a linear order of priority, industrial manufacturing, oil and gas, commerce, homes, automotive fleet, building automation, smart cities, healthcare. These are prime candidates for taking advantage of the Edge computing explosion. And keep in mind, Edge from a market size perspective is four times that of cloud computing. So I always joke, if you miss the cloud revolution, don't worry, Edge is four times bigger. So keep going. All right. So now let's double click on what these specific Edge looks like. And I'm very pleased to report that we have a project called State of the Edge, which is a top-level project under the Linux Foundation Edge umbrella. And that has several of these underlying project on terminology, glossary, etc., which are issued to the community in a Wikipedia-style definition. So if you don't like a term, do a pull request and help us modify it. But this is a neutral, non-vendor, biased terminology that State of the Edge and glossary project has created. And this is the latest publication, at least in terms of a white paper that is being shown here, which goes and defines the terminology of an Edge. So at the macro level, the community is saying, hey, there's really two macro-level types of names. There's the user Edge, and then there's the service provider Edge, the purple and the greenish teal. And they kind of blend in. So it's not like a hard cutoff. And we all know that in what industries and where. So that's the first thing. Then if you double click on the user Edge, you get into three flavors of that. On the very left extreme bottom of the screen, you will see what is called constrained device Edge. These are devices that are constrained because of location, memory, footprint, hardening, connectivity, et cetera, et cetera. They are microcontroller-based devices. They could be embedded. They could be a whole bunch of devices that again, we're not saying what is right or wrong. We are just saying they are constrained. And there is a very important distinction there because you cannot run the entire Kubernetes or OpenStack or any of these high-level software on these devices. Memory is limited, and you can only do so much, and you can only connect to a certain network and let them perform some level of predictive maintenance and AI on the device. So that's kind of the leftmost. Then in the middle, you have the smart Edge, a smart device Edge, which is really phones or PCs or some sort of a gateway or server, which is reasonably protected physically from a security perspective, semi-secure, I should say, in an area. It could be inside a building, et cetera. And then you have the true building factories, homes, et cetera, where it's called on-prem data center Edge. So these are maybe two to four servers depending on the size of the building and factory, but all these are sort of under the control of the end user, whether it's a person, an organization, or a service provider, it doesn't matter. And then you have the last mile, and we know what that means. It's various forms of connectivity to get into a service provider network. And the service provider Edge then starts off with some sort of a base station or some hard-wired telco exchange point and then gets into a smart central office, right? Aggregation hub, whatever. Up to that point, you are in the 20 millisecond zone. And so that's where you're getting most of your latency and Edge applications. Everything after that, regional data centers, centralized data centers, these are not Edge data centers. These are not Edge compute. So let's be very clear on what is Edge and what is not, okay? And I wanted to spend a little bit more time so that people understand that it's not just physical proximity, but also responsiveness and a whole bunch of mobility that goes with the Edge. And again, these definitions are very, very important because you know, we don't want a pollution of a marketplace where everybody calls everything Edge, okay? But that said, let's get into the killer apps. We did a survey about a couple of years ago, and this is still true, where applications have, you know, a set of needs to use low latency and accelerated processing, right? But what are the technologies that are helping fuel this revolution? And that's what is there at the bottom of the screen. So you have the 5G technology. For those of you who are not familiar with this and non-networking folks, it is all low latency, high bandwidth, like extreme high connectivity, right? And this is huge. And it's not just us watching YouTube on a faster bandwidth. That's not what 5G brings. It brings a whole different set of applications and latency is the key factor there. The applications on the Edge are enabled by what is called microservices, right? So the microservices have also been, you know, utilized here for helping these applications. AI has matured where you can actually do predictive maintenance and analysis at the onsite and on these different edges. Hardware has come a long way with all these acceleration techniques. And then the last piece of the puzzle where you can actually work closely with your service provider to provide on-demand network function virtualization or network function slicing or network slicing, as a lot of people call it, right? All those things enable you to drive these applications. These purple arrows are really showing that these technology are building blocks of various categories of new applications, whether these applications are on the infrastructure side, autonomous devices side, like drones and vehicles, or they are immersive experiences or analytics, okay? So really, it is very powerful. And we're seeing a whole set of applications come up. When you do a survey, obviously, you're going to get results based on who take the survey. But this is a pretty good sample of what are the killer apps? There's no one killer app, okay? But what is important is, and I always try to summarize, you know, what are the killer apps and what are the constitutions of a killer app? It's really anything that is non-traditional video, so things coming from drones or streams or cars or whatever, and connected things that could move, right? It could be, you know, a car that drives itself. It could be video content delivery with 360 video. It could be AR, VR. It could be an automated factory, which is a huge use case that's coming up. It's gaming is big, right? Surveillance is big. So these are killer apps. The industry is still working on markets that are utilizing these killer apps. And so it's getting to a point where we are seeing a lot more support for these applications as we move into the edge computing. All right. So with that market background, I'm going to talk about LF Edge, which is an umbrella within the ecosystem. Now, for those of you on the call, if you are not part of this ecosystem, I would definitely encourage you to join or at least take a look at what you can utilize from it. But if I take the diagram of defining the edge and then add another dimension to it, which is the y-axis, you can see that for each of these edge, whether it's a user edge or a service provider edge, you can either have infrastructure-related things, but you could have application-related software, fairly straightforward hardware up the way to software in terms of applications. And so what we have done is we have almost eight projects now in LF Edge. And most of them are shown here. We just had a recent one, which we need to add. I'll talk about that in a bit. But these projects are highlighted by the size and the shade of the bubble here. So there are two Stage 3 projects, Acreno and Edge X Foundry. And then there are Stage 2 projects, again, which is one step below, which is the home edge, Beetle, Eve. And then we have Stage 1 projects that have a lot of interest and visibility and community. And they'll just progress into the future stages. And that's Fledge and Open Horizon. And then, of course, State of the Edge, as I talked about is Stage 2. These projects are situated in this X-axis where you've got Edge X Foundry very close to the application frameworks for IoT. Fledge more on the constrained environment, if you may, towards more of the left-hand side. Then you have infrastructure projects. And then Acreno itself, there's an arrow going back and forth. That project has two parts to it. One is it brings the telecom or telco use cases and software. But it also provides the framework by which blueprints can be created across the entire LF edge, all the projects, not just LF edge projects, but upstream and downstream projects. And this is the blueprint, which I talked about for the circle, to accelerate deployment. Okay? All right. So let's put this in a physical diagram. And now you will see why Open Source has really, really come a long way to provide a whole bunch of solutions for everybody in their landscape. So starting again from the same thing, you've got the left-hand side, the mobile networks on the top, left, residential, small, medium business, and enterprise, whether they have the on-prem data center or gateways, it doesn't matter. Then you have the edge coming in and then you have the cloud. And if you look at the projects outside LF edge that will collaborate very closely with the LF edge, you have what is called O-RAN or Open RAN. This is a telecom initiative for open radio access networks. They will be collaborating, in fact, they already have with Acreno and Stack from a blueprint perspective. And then you have sort of the enterprise side of things with EdgeX Foundry, Fledge, and even Home Edge. Are we good? Okay. I'm ready to go. Ready? And this really shows that it is live. Okay. And yes, I don't have a IoT gateway sitting in my basement here, which I don't have recently, it's California. But anyway, so let's go on. It's the second part of the presentation, obviously. So we'll have two flavors here. So anyway, so what I was saying was we have the consortiums and standards. So Etsy Mech, a big participant here. And then we have ACC and we have IIC that is important for this. So let's just go quickly into the next slide. And these are some of the use cases that are coming up on the LF Edge projects. I think if you can hear some feedback, I could probably microphone on my hold on voice is still going through the web pass. So if I mute the, okay. Can you hear me now? If I mute the phone, how will it go through the phone? Sorry. Okay. I'm just going to continue with some feedback. I got a text saying we can hear you. All right. So the use cases here are on the home and side anomaly detection surveillance. And I'll talk about that in a bit. From the telco edge, there are blueprints around radio cloud. There's blueprints around edge applications for connected vehicles, connected classrooms, etc. Then you have building automation, controlled smart cities. And then, of course, there is the IOT predictive maintenance. So we'll talk about that each of these projects and their blueprints very, very quickly. Before we get to each of these projects, I want to make sure that we understand that there's a landscape like all our Thank you. You will now I'm going to talk about the LF Edge in terms of summary. Okay. There you go. So LF Edge, again, our vision and our goal is to unify the frameworks that the IOT, the telco edge, the cloud edge, and the enterprise edge create. And clearly, we mentioned the project, but these frameworks, because each of these frameworks, and we call it the plumbing layer, are important for a lot of the cross-section of the population. If you look at the members that join and participate and influence, they are from a wide cross-section. They are all different types of hardware and silicon, because it's independent arm, Intel, Qualcomm, etc. You've got the cloud players, Baidu, Tencent, etc. You've got the industrial player in a dynamic, OSISOC, etc. And then you have a whole bunch of telcos and other participants that are looking at not just collaborating, but also participating in this. So let me see if I can go to the next slide. All right. So here we go. Okay. There you go. So the edge community is thriving, and it is very lively. Okay. And you can see that there's been a huge member increase. There's been a huge project increase, lots of deployments going on, downloads. I mean, even this slide is out of time. Okay. So we can't hear at all, or maybe not. Okay. I don't know. We can hear you. We can hear you. Oh, we can. Okay. Okay. Then that's fine. Okay. Because I was getting no feedback at all. Okay. Then I'm good to go. All right. Okay. So there was a heavy increase in the developers. So one thing I also mentioned is during the pandemic, which we are all going through, work and coding is the best form of distraction for a lot of us, right? So what's happening is we're seeing a steady increase in the number of developers as we don't travel and as we don't do a lot of things that we normally would do. So thank you again for the community that has been participating. You're creating huge value because if you look at one of the Forbes articles, it said very clearly that IoT and Edge are the most important things post-pandemic and we better get ourselves ready for that. So anyway, so the community is going significantly higher and going quite well. And so now we are going to get to the introduction of LF Edge projects. And let me go into the specifics of each of these projects. As I said at Crano, it is aimed at making sure that telecom use cases and cloud services at the Edge operate seamlessly. And it's doing it through a set of blueprints. Edge Expoundry, and by the way, these are stage three impact projects. They've been around for more than three years, a huge community of following very solid deployments, very solid downloads, right? Edge Exis, the IoT framework that I talked about, the plumbing layer. And we'll go into each of these in detail. Eve is an on-prem project, on-prem virtualization project, edge virtualization, contributed by a startup called Zedda, but now it's gotten a community behind it, Home Edge, again, contributed by Samsung. And it's getting to a lot of real releases that are coming out. We'll talk about that state of the edge I already mentioned. And then we have stage one at large projects that are gaining a lot of momentum. For example, Fledge, open source for industrial edge, specifically architected for IoT. And again, remember that constraint device picture. Beedle, again, it's coming from the cloud, and how do you do a general-purpose cloud platform, life cycle management of APIs and things like that. And our newest project, Open Horizon, which came from IBM, is really managing the life cycle of containerized workloads for the Edge, and including machine learning. So very, very good-flavored projects that are there. So let me go through this very, very quickly. And I know there are a lot of questions, so I do want to leave a lot of time at the end for answering them. Let's start off with Ukraine. The diagram here is just a deeper diagram of the first one, so we won't go into that. But what we have here are blueprints, and right now there's over 16 blueprints. And blueprints are something that the community has tested in a declarative configuration that you can actually, you as an end user, can actually utilize and rest assured that the entire software and hardware and declarative configuration of that particular use case is good to go. So you have, and here are just some examples. So if I want to take software and create a cloud at the edge of the data center or below a basement, whether I have a one server or six servers, that's the network cloud blueprint or the radio edge blueprint. There are blueprints on the Mac side of things. There are blueprints that talk about connected vehicles, right? How does a vehicle connect to a base station into a cloud? And how do you do more precise use cases beyond just a GPS, right? That's that blueprint. You have a huge blueprint that is being very actively used right now on AIML, as well as augmented virtual reality at the edge. One example is the classroom and how a lot of users are monitoring through AR VR and applications on LF engine, specifically cranial, the classroom experience of students, right? In a virtual world, critically important, there's blueprints on automated factory. So keep in mind, what this, what the goal here is it's a finite set of configurations to reduce complexity. We've all coded and we've all been, you know, engineers in our lifetime where we would have, you know, enter parameters one through 25. And then there is another tree that you go and enter another 50 parameters. And then you, before you know it, you have an end buy in combinations of testing that you got to do. At the end of the day, majority of them are not needed. So let's go with a finite set of configurations to reduce complexities, which is what defines a blueprint. And some of these blueprints are really optimized for the edge. Okay. And these are turnkey solutions and very, very important projects. So if you're not part of a cranial, please join. Okay. The second project is edX Foundry. Again, these are in just alphabetical order, but more importantly, stage three. edX Foundry is one of our older projects where it creates a very flexible IoT framework. Okay. Not everybody needs to abstract the physical devices. And not everybody needs to figure out the 160 messaging protocols that go back and forth. And not everybody needs to figure out how to connect to the cloud. So the right small diagram is the architecture diagram. And really it's all around, you know, how do you really take advantage of a generic plumbing layer or framework so that you can focus on your value add and not just on connecting and lifecycle management and things like that. So edX Foundry has been downloaded 5 million. Now it's probably 6 million when the slide was created. So very, very important project for us. The use cases that I want to highlight are all focused around manufacturing, for example, where, you know, you could have remote monitoring of production equipment, et cetera, et cetera. Then there's the retail market, which is a very big user of edX, edX edX Foundry, you know, interchangeable nerve terms. Here, you know, I know retail is right now, you know, in a different situation, but it's even more important to have much more of a newer experience on retail. So, you know, we're seeing a lot of use cases and a lot of the requirements getting added there. And then, of course, building automation, smart buildings, controlling all of that, you know, sitting on an IoT gateway. Originally, this came from Dell, but I think now it's a very diverse community, you know, very important part of the ecosystem. Now, the key point here is this is agnostic to the hardware, agnostic to the operating system, so it doesn't matter what OS you load on it. It is agnostic to the protocols, sensors, and the endpoint. And I'm going to add, you know, independent of the language in which you write, okay? And it's just written as a loosely coupled, distributable set of microservices. So, think of it, it's a loose collection, so you can actually bring it together very, very quickly. Okay? So, that's Agax. And then, if you go to some of the stage two projects, we have Eve, that is on-prem, and it's an edge virtualization engine, as the acronym would say. But it is used to simplify the development and orchestration and security of a lot of these applications that sit on-prem, right? Where connectivity is not guaranteed, it could be intermittent, it could be permanent, it could be partial. So, how do you consolidate a mix of, you know, VMs and containers, maybe legacy apps on an edge hardware, right? That's sitting somewhere in the on-prem area. And how do you do it securely? Because the parameter is no longer there. There's a lot of focus on the zero trust security in Eve. There's a lot of support for bare metal and orchestration, right? So, keep in mind, this project is very valuable on a whole bunch of on-prem gear for IoT, okay? And it kind of feeds into the rest of the projects as well. So, that's Eve. Then you have the home edge. So, think of that equivalent inside our homes, right? Where you have a whole bunch of different connectivity options, right? Whether it's your TV or your fridge or your Nest or your thermostat or, you know, those are one thing. But then you have more two-way interactions, more services coming through, more surveillance coming through and so this project is particularly focused on, you know, creating that life cycle management robust framework for the home edge, right? For the device, the software that goes on your device inside the home. Specific use cases that have already been sort of released in the open is service offloading, right? When a device doesn't have the capabilities, you know, so move things off. No latency, compute frameworks for a lot of real-time data, etc. How do you do device management? How do you do service management? How do you do, you know, which device performs with service, right? So, there's a lot of cool features that are being developed as part of home edge. State of the Edge, again, this is a non-software project. So, as I said, it produces research, it produces glossary, it produces landscape, it produces a lot of important things that the community can all align on in terms of definition, etc. and make sure that edges is a location, not a thing. There's a lots of edges, but at least we're trying to refine and define it properly, etc. So, in April, the whole State of the Edge moved in to LF Edge as a top-level project, which then folded in the glossary and things like that. So, a very important project here. And then, of course, some of the very exciting new projects that have really taken off quite well. Betel is one of the projects where, you know, coming from a cloud side of the world and coming from an edge, how do you blend that line so that you can have a lifecycle management of applications on board, right? You can process drones, for example. And how do you go and put AI ML as close to the edge as possible so that, you know, you can inspect images, video images via AI of a drone, okay? I know this sounds like very, very futuristic, but it is here today, right? Releases are there. And you can actually, it works on x86, ARMS, MIPS, it's OS agnostic. So, you can see the theme of the projects we have. OS agnostic, hardware agnostic, cloud agnostic, connectivity agnostic, etc. Because those are the frameworks where you don't want every company, every product, every application to reinvent the wheel. That's what we do. That's open source, okay? That's not where you differentiate. So, let's all come together and not reinvent these basics of edge. And then, as I said, Fledge in an industrial environment, IOT, constrained community, you have a framework here, which again has several of these use cases, whether it's, you know, transformers or turbines or, you know, any of these mines and factories where predictive maintenance and AI analytics is needed. We just released the TensorFlow support on it for Google on the north side of things. Plus, it's got, you know, integrated security management and things like that. So, again, very, very exciting projects here. And then the newest project that joined LF edge is open horizon, where, you know, you're managing life cycle and a containerized asset, including machine learning asset, and then distributing it into these environments, right, in a very high scale. So, very important on this one. Now, keep in mind, on surface, they're all IOT projects. On surface, they're all edge projects. On surface, it may look like, oh, my God, why so many projects? But that's where blueprints come in. That's where a crane all comes together and brings everything together. Okay. So, key takeaways. We are here to harmonize and unify these various edges that IOT enterprise cloud and telecom see, we are here to keep the edge open and interoperable in collaboration with consortiums and SEOs. And it's similar to other huge projects like CNCF and LF networking and things like that. If you want to get involved, just very simple. Step one, get an analytics foundation ID, visit the wiki and start participating in any of these things. Okay. And that's your wiki for the edge. I think we have about five minutes or so to take questions. So, I'm just going to go in a linear order. And obviously, the first one was audio. So, that one's gone. All right. Okay. These are all audio issues. Okay. All right. So, the first one was what the purple are. Okay. So, that one's okay. I think I answered that. All right. Okay. If you have any. Okay. All right. Okay. So, I think one of the question is, how do you work closely with standards and other organizations? Okay. So, clearly, I think I mentioned that SCMAC is important, IAC and ACC. So, ACC is the Automotive Edge Computing Framework, our consortium, which is the Toyota's lead edge computing for automobiles. So, we work, they are an associate member and, you know, blueprints are created in Crano for that. So, that's kind of how it works. If you have any other questions, please see if you want to type it in. I will be able to answer them in the time frame if we can get through all of these. Okay. Next question. Okay. All right. I think most of these are audio things. Anyway. All right. I think that's the end of the presentation. I'm going to stop now. Thank you for coming. Sorry for the audio issues. I think we had tested the system, but not sure what happened. Again, keep the conversation going. Visit the number two track of the project updates and keep going. Thank you very much.