 So thank you, good morning. So it's an exciting area topic, EDGE. What is the EDGE? As Arvid said, everybody has a slightly different definition. So Arvid and us got together to do this research. And you can't go out and say, what are all the EDGE applications today in motion. We talked to 14 experts, 14 thought leaders in this space and got their take on where EDGE is going. Or more true, I think, would be to say the EDGE is. Everybody has a slightly different take on it. So just quickly, a lot of people know me as from Infinetics Research co-founder. But we're part of this big company, IHS market. 50,000 customers worldwide, 140 countries. We really have a full spectrum. And Infinetics fits into the technology group where we track components, materials, the devices and equipment, and then the end markets. In my case, it's been telecom, operators, et cetera. And we have industry impact, 127,000 published articles with our researcher analysts quoted. So what we do, I call it the embryonic EDGE. There's a lot of EDGE applications in motion already today, but what's coming? Where's the revenue opportunity? Where's the cost? Where are the impediments? So we're going to touch all that. So we talked to thought leaders from major telcos, MSOs, equipment manufacturers, chip manufacturers, startups, some key startups, open source, and spread around the world, Asia, North America, and Europe. We defined the EDGE as a place where compute is located. That's within 20 milliseconds of the end user device or computer or whatever it is that's accessing the network. And so 20 milliseconds, what does that look like? So if we look at a map, if you will, of a network, we have the EDGE on this side, and back and deep in the network where the data centers are, big data centers. So 20 milliseconds, actually, here is from the EDGE device to the COs, anyway, and to some Metro IXs as well, some not, though. OK, so that's where we define it is. So at the EDGE, well, certainly IoT, gateways, what's not an EDGE, a big data center. What is an EDGE? Partial EDGE, a regional CO, or certainly, I know in the San Francisco Bay Area where I am, Equinex has at least six locations around the Bay Area. And in each one of those, there's some distributed data center from Google, from AWS, Microsoft, et cetera. So that's a partial EDGE, I see, because within that 20 milliseconds, you can reach a lot of the population, but not all of it. The CO, yes. So the stats I'm showing you here are from our big study we did last year with operators around the world that represent 61% of the world's CAPEX. And they said that where are they going to execute BNF? So that's one of the functions classically targeted for EDGE compute. So we see 97% of the operators at some point are going to put BNF execution. That is, they'll have a smart CO. They'll have a data center, a small data center, in that CO. What's another place? Well, over 80% of the operators say they're on these new devices, a universal CPE, which I define as a little server that has a Pico Cloud in it, sitting on an enterprise location. That's an EDGE, too. We do compute there easily. And then in homes as well, coming along soon. That also certainly at its cell towers. We talk about 5G, but even 4G, there's a lot of applications here already available. Are targeted anyway. MEC, AI, the IoT, and certainly 5G when it gets there. And the EDGE certainly includes the COs and some of the AIXs. OK, so now I'm going to talk to you about some of the data that we got from our interviews. My colleague, Dr. Cliff Grossner, and I did all the interviews. We did them all live, took a lot of notes, and came up with synthesizing this into some actionable data for you. So there's a lot of data. It's downloadable. The file's there, so I'm not going to go through all of it. But I want to at least give you some of the highlights. So if we look at the top five applications, and this is by the percent of respondents, video content delivery and video associated applications are prime number one. And Cisco VNI predicts that in 2021 that 82% of the world's traffic will be video. So video is a huge, important application, or just the bandwidth that's being used with video needs to be solved by putting content delivery, networking, putting caching, and some other applications at the edge. So you'd reduce the bandwidth, obviously. Autonomous vehicles or drones or fleet management, et cetera, is the second one there. ARBR IoT was mentioned by a lot of people, especially factory IoT. So if we look at those IX locations where Google, et cetera, is in a Bay Area, for example, but a lot of other metros around the world, they have access to part of the population, but they don't have access actually from the edge. They cannot achieve the delay for a lot of edge applications because they don't have the locations that are close facilities. OK, so revenue, so these thought leaders thought that there's five edge applications. And let me say that our edge applications, each one of them is actually a collection of applications. So there's five that got the top rating for revenue production. And so some of them are consumer-driven, like gaming and video CDN. Others, the enterprise-driven revenue, private LTE, industrial IoT, supply chain management, et cetera. There's some non-revenue-driven, like surveillance in smart cities, security and safety as a driver, or smart cities have their own reasons to deploy welcoming environment for businesses and for their citizens. So let me move along because I don't have a lot of time. I'm going to be around, and also you can always email myself or click grocery with questions. The top cost technology, sorry, the top hurdles are technology and high cost, pretty generic. But how do you manage and monitor and operate and control hundreds of thousands or tens of thousands of edge applications? Big issue, AI is involved. We're going to be involved, but AI needs to be trained. You have to have big data to train it. But what do you put out at the edge? Not big data, you put an inference engine. So big problem is there's no common APIs for developers to develop edge applications. There's a lot of experimentation to figure out the business cases. So here's an interesting one where we have what areas need to be invested in in order to fully develop the edge? Well, on a scale of one to seven, we always, in our methodology for research, take a six or a seven as those are critical. Those are really heavy answers. So what do we see? Oh, look, there's 10. There's 10 areas that are significant areas for investment to fully achieve edge compute. So just fascinating. So there were a number of direct quotes from the people that we interviewed. I'm going to start from the very bottom. There's a lot of good ones here, but sell site with E-node Bs and 5G. Nice latency, good location, but huge dollars to deploy everywhere. And you must have business cases. The next one from the bottom on the right side, the COs are not designed to be data centers. There's a tremendous challenge for the hundreds, thousands, tens of thousands of distributed edge cloud, the administration, the management, OSS-VSS connections, distributed database. There's a lot of issues to solve. And so the five top takeaways, the top five edge applications are revenue, industrial IoT, video CDN, distributed and virtualized mobile core and fixed access, private LTE and gaming. And we noted that the ICPs and OTTs have partial coverage with distributed data centers and big metros, 20 to 50 milliseconds to the end user devices. But telcos can offer much closer 5 to 20 milliseconds. Talking to one operator this morning, who you'll hear about later this morning, hear from, their targets under 20. So we learned that a lot of the edge deployments are justified first by cost savings and then followed by let's deploy revenue bearing applications there. A lot of it makes sense, just like SDN and NFV starts small. Learn about the operations, the business cases, and the difficulties involved before you start expanding those edge applications. A lot of the applications are going to be deployed soon. And many of those are going to get enhanced over time as the edge expands. And then finally, edge compute is complex. And you have a lot of things that are going to be placed at that edge compute locations. NFV, MEC, fixed mobile convergence. And how do you manage the 1 to 10s of thousands of locations? You have AI inference engine, security, distributed node, data consistency, a 15-year lifetime. So there's a lot to go on. Here's some new fresh data about what the revenue bearing apps are, what the important apps are going to be. And if you have any questions, please feel free to email myself or Cliff. And we're happy to answer them. And I'll be around to engage in conversations today and tomorrow. Thank you.