What use cases does Edge computing enable?

Question

Answer

Edge drives home the point (probably better than any other technology) of value and value creation. If you look at typical supply chain models and typical value chain models, in a value chain model you start at the customer and work your way back to materials. In a supply chain model, you start at materials and work your way over to the customer. We live in an age where the customer has to be part of the criteria of any technology purchase that an organization makes. And I believe that Edge brings that value proposition of customer centricity and customer first, because it’s about two things: shortening time to value and accelerating the pace at which value is created for your customer, employee, supplier, and also for your ecosystem of trading partners.

Think about VR or AR as one little example. I can use my augmented reality to troubleshoot a line in a facility on the other side of the globe. My background is in large enterprise, and I've dealt with situations where there are 34 operating companies and the biggest networking hurdle you have to deal with is, "How do I get information over a mountain?" I have two factories both in Mexico City but there's a mountain in between. Edge allows me to create some of the value on one side, and get the rest of the value from the other side. But because I have something either locally oriented towards me, as opposed to Cloud, or in an Edge environment, then I can use that as a pass-through without having to jump through hoops to bounce multiplex signals off of a satellite just to get over the other side of the mountain. So because I'm able to do that, whether it's with AR or VR or any other form of technology, I can troubleshoot and use hands on the other side of the mountain to fix the device that's causing my production line to slow down.

Answer

I look at Edge computing from the perspective of the flow of information and activity. One of the big things that we're seeing is actually a sort of natural evolution. Geoffrey West wrote a book about scale, he's out of the Santa Fe Institute and studies complex adaptive systems. It turns out...when you have a system built to optimize the flow of something, things consistently tend to evolve in this sort of trunk, limb, branch, leaf mentality of breakdown. Look at the way your circulatory system works, the way a city organizes traffic, the way that commerce moves through shipping. This really consistent pattern is something that I think with computing... we've sort of been doing but not very well. I believe very strongly that we're sending too much back and forth between central data centers and wherever the computing's being done. It's getting to the point where the amount of flow that we have between those points is just untenable for that kind of distance. So we need something where we can do more of the computing at the location or in the group of systems that are related to each other, before we send more packaged data/centralized data through the pipe to the more central systems that are doing large scale analytics, building training models to send back out to the Edge, whatever it may be.

So one of the things that we're definitely seeing is this idea that we don't want a whole other infrastructure. We don't want a whole other experience to build applications and to run applications on those Edge locations that we do in the central environment. That's not the point. We do know there's differences, we do know there's different requirements from an operational perspective and otherwise. But ideally what we would like is the ability to say, "Hey, we're going to build an application. When we build that application, if we choose to deploy it at the Edge, it's just the same mechanism to do that as to deploy it on one of our servers in our data center." So for certain classes of applications that are very consistently built (e.g., web applications, mobile applications, event driven applications) there should be commoditization of underlying infrastructure and underlying interconnectivity, but with a lot of flexibility for what the developers can do with it (e.g., a lot of customization that can be done on top of those base elements). So delivering tools to build consistent paths to production for different organizations for different purposes for different needs is a big part of what we see happening.

I do think things are going to change, though. I do think we're dealing with a model today that's still leaning back and putting its weight on the client server model a little bit. If you begin to open up the ability to just to connect to event streams really easily and lower the cost of integration with real time data, you're going to see this explosion of capability of sharing activity across organizational boundaries. Instead of outsourcing an entire process, we’ll be able to outsource a step in a process. We’ll be able to do things at a different scale and in a different way than we did before and that, I think, is going to fundamentally drive really cool new technologies. Things like that nats.io IO folks are doing with the Synadia, trying to build a whole dial tone for interconnection through messaging and event driven systems. I love what EDJX is doing. I think it's a really, really interesting play. I think there's a lot of really powerful things to look at there. So I do believe that a lot of this stuff will get hidden from developers having to directly interact with it and directly build with it. There'll be a lot of serverless concepts on top of it. The opportunity with Edge is really about changing the data gravity of these applications and operations that they have that are spread so geographically diversely in terms of the application subsystems that are doing various actions.

Answer

I think that what makes new technology fascinating and interesting and important is really the root of it, which is challenging existing paradigms and existing modes of operation. 'We've always done it that way so we're going to do it this way again,' kind of approach. I think that's what motivates folks like me and James and Mark, in the sense that we try to look at the world through a different optics or a different lens and figure out how it could be done differently as the needs and demands change. That's really what is fundamental to understanding the Edge, that it's a new dynamic. It's a new paradigm in the sense that we're moving away from a world where we built for web services, we built for mobile, we built applications in a centralized, monolithic way, we programmed to a central core and that's what we did. What necessitates change at the Edge is that that paradigm has to break. It has to break because I can't possibly orchestrate monolithic applications to all the millions of locations that it may be necessary to run or operate that application. I've got to break it apart, which is why we arrived at serverless. We tried to do this with orchestrators and with Kubernetes and Docker. We could make it work in a single instance but we couldn't make it scale to end. We couldn't do it. It wouldn't work. The reality is that, to James' point, what is a unique characteristic about applications in the Internet of Things is that they are rooted in event driven code. We call them nested functions. You can spawn functions on top of functions and the application in that kind of fashion exists everywhere. It doesn't exist in one place. It delivers a thing, it delivers a service, it delivers value…. but it does so in a highly distributed or decentralized way.

What use cases does Edge computing enable?

Content you might like

Has anyone found a AI solution to screen IT contracts to reduce the time it takes to onboard a new vendor solution?

Do you expect cryptocurrencies to be more or less valuable at this time next year?

Have you modernized your Problem Management process to include machine learning/AI capabilities in identifying problems and potential solutions? Taken IT Problem Management beyond IT into the business processes? Merged SRE practices into ITIL Problem Managment?

What's a quick win for optimizing spend? What's been your focus or "lowest hanging fruit"?

Do you agree with this statement? By 2025 Cyber Attackers Will Have Weaponized Operational Technology Environments to Successfully Harm or Kill Humans.