 Thanks everyone who made it back here after lunch. I hope your coffee is already kicking in and I will also do my best to do at least a half of a good job as Amar did with keeping time. So today I've prepared you some snippets in terms of edge challenges and a solution example. So let me just dive in. Before getting to the content, I'm Yelda Govancha. I work as senior manager of community and ecosystem at the Open. I'm an infrastructure foundation. Now you can really hear me and I can breathe. So if you did not, would not get a chance to talk at the event, you can find me an email or ping me on Twitter or any other chat or social media platform where I'm available. I would love to talk to you about edge computing and to give some ideas in terms of topics that topics to cover. I prepared some food for thought and in the first half of the half of the talk. So use case examples. On your left side the diagram and the text under Instal communications and 5G and you probably ask why it's even on the slide, right? I like to keep it because on one hand it is a good example for something that is running in production already and if you ask me on my phone, it doesn't always work very well. But it is running in production and if you check the diagram on again your left, then you can see that it is something that goes out from the core to the edge and in my experience when people talk about edge computing we often tend to focus only on the tiny edge and forget about what it is on the edge of that it belongs to usually a much larger connected mess of the distributed system that in fact provides you with a lot of challenges and even if you only look at the right side of the diagram with the cell towers and user and machine devices that are connected that still is a challenging area. On your right side, I brought another example which is industrial IoT where we are currently in the stage of factories getting digitalized and they are looking into how to apply clouds on the factory floors, how to utilize a better way of again using their resources and at the same time dealing with challenges like mission-critical applications and safety of not just machines but also humans. So how you deal with a lot of sensors and components on one factory floor and how you deal with multiple factories being connected and if you look at the text on the slide you will see a lot of similar characteristics and requirements in terms of hardware acceleration. Again, some real-time operation in there. What happens if you have disruption in your network? So at the end of the day these two use cases might not be that different from each other, even though one is in some extent much further ahead than the other one. So as a summary, I have a slide about production challenges, but what I really want to focus on is complexity because the other thing that I hear really often is how we can come up with something simple, a simple solution, but at the end of the day, especially if you think about the telecom example with the massively distributed large telecom backbone infrastructure, how can that be simple? Does anyone think that it can be simple? Don't see hands yet, so probably not, but you can always come and prove me wrong. So one thing that I hear more and more as a conversation topic in the area of edge is how important automation and orchestration is. So that is something on this slide that I really, really wanted to highlight. And we have some solutions in that area, but I think that it's something that we can all work together on coming up with even more components in that area. And to the end of this segment, I prepared some questions and I don't have answers for you because most of these questions, in my opinion, will have different answers depending on what your use case is, what your solution is, what your interest in edge computing is. So, for instance, what is edge computing? Is it only at the edge, that tiny edge side, for instance, or is that the whole core to edge infrastructure depends on where you sit and where you're looking at the system from? Or another example of conversations that I've been hearing recently, for instance, a comment that said that Kubernetes was designed for large data centers and is not applicable at the edge because of, again, the structure of the edge and the requirements at the edge. What do you think? Do you think it's true? Do you think it has a point? Do you think it's absolute nonsense? Who knows? I think it's a great conversation topic. And, again, I personally don't have an answer for you that would rule it all, but I would love to chat about these questions and some other ones with you when there is some time for that. And just to show one example that kind of goes into an aspect of some of those questions and some of the requirements and complexity and automation, it's the Starling X open source project. It is a fully integrated open source edge cloud platform. So, in that sense, on the previous diagram, it does have the distributed infrastructure model and it's something that you would use and install that would cover your infrastructure from the core to the edge. That's the context and that is the scope of the project. And when it comes to Kubernetes and the CNCF ecosystem, you will see that there's Kubernetes in there. There's flux seeding there, Helm containers. So it is a project that integrates Kubernetes and it adds some more components like the purple icons on the diagram that are helping in the area of that automation and orchestration to manage that end-to-end core-to-edge infrastructure and provide, again, an end-to-end home for your edge applications. I'm probably running lowish on time, but still all right. So when it comes to, again, something similar to what the Telcom and 5G use case diagram was showing you, what Starling X covers is you can install a central component in the central cloud and have your central cloud infrastructure covered with a single pane of glass management dashboard and you can use that dashboard to deploy edge sites on various sizes on one server, multiple servers, or like a regional data center-sized environment, different high availability scenarios. And what's cool about the project is you can have a small edge site with only container workloads and what the platform takes care of for you is keeping all the sites in sync with the central cloud and making sure that if you have disruption in the network and an edge site loses connection to the central cloud, then once the connection builds back up, it becomes synchronized all over again. And in the meantime, the edge site does have autonomy, so it still has kind of the full control and operation available at the edge site. And when it comes to the one single server deployment, it is just to use the fancy buzzword hyperconverged. So you have all compute storage and networking functionality on that one, sometimes tiny server. The latest release is 7.0. There are some examples on the slide of some of the features that I knew. What I would like to highlight is things like the precision time protocol, which is something that is usable both in the 5G use case as well as in something like industrial IoT, where, again, those real-time mission-critical applications are highly dependent on PTP and also time-sensitive networking. And it's an open-source project, so if anyone is interested in checking it out, just a few links on the slide in terms of where information and the community are available. And if you would like to meet the community, then we do have something that we call opening for life. It is live, one-hour-long interactive show on Thursdays, so you can catch industry experts and community members talking about open-source infrastructure, challenges, solutions, predictions for the mid-long-term future. And if you have any ideas in terms of what you would like to talk about, please submit them on ideas.openinfra.live. And if you would like to see the community in person, then the next Open Infrastructure Summit will be in Vancouver in June next year, and our CFP opens in just a few weeks mid-November, so I hope that we will be seeing you there. And with that, that's all I had for today. Just in time. Thank you.