Elevate the Edge
Elevate the Edge
Ian Hood, Chief Strategist at Red Hat Discusses Cloud RAN in Retail, Manufacturing and Energy
Ian Hood, Chief Strategist at Red Hat Discusses Cloud RAN in Retail, Manufacturing and Energy
Hello and welcome back to elevate the edge on Maribel, Lopez, I'm joined with my fabulous co host, Jo Peterson. Hey, Joe. Hey Maribel. Joe is wearing a beautiful pair of turquoise glasses, I'm gonna have to start taking photos of these so that I can put them on the internet and everybody can see how cool her lens where is everyone lovely. We're also as always excited to be talking to another great leader in technology. Today, we have Ian hood with us, and is the chief strategist for global industries at Red Hat. This means he acts as a strategic adviser engaging Red Hat's global industry community of customers and partners. He's focused on enabling innovative edge services to improve their products and services, including accelerating the delivery of innovative applications across distributed edge clouds. And you know how much we love talking about the edge, you know how much we love talking about cloud. Today, we thought we might talk a little bit about cloud run and retail manufacturing and energy. You know, I've spent many years following the telecom and networking industries. I've been following along with Joe some parallel trends, such as cloud native applications, as well as a disaggregation of functions for radio access networks. And I'm just jumping in with a question before even welcoming and so I guess, perhaps I should stop and say, Hey, and welcome to the program.
Ian Hood:Well, thanks Maribel for having me. Thanks, Joe, for inviting me, always fun to have these these discussions and talk about what's going on in the industry. Because, you know, we were just a mobile Congress just last week, and lots of buzz in this entire topic. So good, good timing.
Maribel Lopez:Yeah, I'm really excited about it. Because I think we've been talking about things like, you know, open radio access networks, we've been talking about moving applications to the cloud, you know, both of these platform technologies appear to be mature enough to converge in a production environment. So given that we're just off of the off of Mobile World Congress recently, how do you see this convergence, unlocking new opportunities for communication service providers? I guess, maybe we can just start with some context.
Ian Hood:Question. And I think the easiest way to kind of look at this one is to take it down the open platform, right? Right. Because if you've got a common platform, you improve operational efficiency, we can go after sustainability and security. And then we'd give our operators choice of their network and edge application software suppliers, right. So a lot of choice giving them operational capabilities. And what we find is that if we go that same platform from the core Taran, we can actually use that same platform to deliver those edge services we talked about. And what's going on is that, you know, the AI and analytics, you know, taking that advantage of the gold, it's known as data inside the industrial manufacturing oil and gas utilities, either at the telco cloud edge, or right at the enterprise edge, lets them go after those, those types of applications for those types of customers. And what we found is that in many industries, and now we're going to apply it to telco just like we typically do, we go from sort of the industry side of things, take it to the telco. And we found that it is proven to accelerate innovation across these enterprise customers, whether it's banking, oil, and gas and automotive, just as we were talking about. And now we can do it with the operators themselves. And this accelerates the time to value. And also, in some cases, we've seen it possible to actually enable new applications or services and environments in minutes.
Maribel Lopez:minutes. I mean, it's just such a different thing compared to what we thought about before in terms of how long it took to stood up, stand up infrastructure and actually get something deployed. You know, I know you've been collaborating with telcos to get cloud ramp solutions ready for live networks. You know, Red Hat's been doing a lot of work in this space. What some of that work look like today, you know, what are some of the things that we might be doing that are different than, you know, this more standardized, long, slow moving processes we've done in the past that we closed?
Ian Hood:I think the thing here is that we've been working in tandem with both our service provider customers, as well as the RAND ecosystem to actually mature and harden and automate these deployments at scale. And we've established partnerships with the key players, you know, the Erickson's Nokia, Samsung's, mavenir, NEC, but we've also had established partnerships with the silicon players, Intel, Nvidia Qualcomm and Marvell because there's a bunch of current requirements for ran. Now, we got latency to deal with, we got performance, and what other ones people tend to forget about, which is if you want to actually have a nice, clean, crisp voice call, right, you need to synchronize the network entirely. And that takes a lot of hard work. We've done this over time in this category, so we're working to actually improve that as well as to deal with power consumption to help operators, you know, sort of meet those sustainability goals. And one of the things as well though, is that we've taken what we call Kubernetes opera routers, it's a common way to actually control and configure the network and CRDs. Those are custom requirement capabilities to enable that consistent instantiation of the ran workloads across all these different hardware platforms, as well as the declare the what we call the declarative get ops principles to enable this zero touch approach to automation and provisioning at scale. Right. So lots of work and lots of words.
Maribel Lopez:No, but I think you can underestimate the coordination effort that needs to happen for this to all work and that it is an ecosystem. It's not any one single vendor single technology that could make this type of thing happen. Over to you, Joe. Definitely not.
Jo Peterson:So in case folks are curious, what does Red Hat OpenShift have to do with Cloud ramp? Can you explain the connection there?
Ian Hood:Sure. And so basically, I talked about the platform. So Red Hat OpenShift, is that consistent cloud native app development and deployment platform, so I can develop apps on it, or I can deploy them. And I use it in public and private clouds. And we can use it for 5g core. And we've actually scaled it down to a single node version that can support these random deployments at scale with minimal overhead. So what we do is we take this OpenShift platform, we tune it for the random type of workloads. And then we've worked with our partners. So we can actually onboard these applications, operate them and make sure they meet the performance of the three pieces of the ran in this world, which is the centralized unit, the distributed unit, and the radio interface controller, those are three pieces of what's known as the cloud ran. And those software applications are on boarded onto OpenShift at the edge of the network. And then the other thing that we do is because we're you know, we are the open source company, we also work in the open round community and many other open source communities, to ensure that our partners take any of these innovations that we've come up with together and bring them back to the community. So we can accelerate innovation across the industry. So it's kind of a two part thing. We do the innovation together, but we got to share it with the rest of the world.
Maribel Lopez:Yeah, that makes a lot of sense to me. You know, we've, yeah, I think we've been talking a lot about the service provider landscape. And Joe is always great about finding these statistics. So she came up with the statistics that said that the cloud radio access network is expected to generate 70.4 6 billion in revenue, I like the four six part by 2030. So there's clearly some mad skills model going in the background and excel on that one. And that enterprises account for three fifths of that market today. So you know, I started the conversation with saying we were going to talk about some other industry verticals as well, maybe retail manufacturing, can you share how you're seeing some of these industry verticals, maybe even energy utilizing?
Ian Hood:So a really interesting kind of way to look at it. And, you know, those statistics are kind of interesting to me as well, because the majority of kind of how many of us is going to use ran, and is just like we all do today, it's at the end of our mobile phone. Right. But there will be, you know, 5g radios that sit out at enterprises and just basically connect them up over the, over the air to something else. But I look at this and say the cloud Ran is really more an approach for the operator to expand their 5g network to adjacent markets to go get more coverage, you know, be more competitive. But the second thing is that they actually can go offer these Mechon enterprise edge services. And that's really kind of what it allows them to do when I'm using an open platform like Cloud ran. Now, if I take a look inside what's going on, really, at the enterprise, it's really more of a private wireless 5g or LTE network that includes this common cloud application platform for both the core and the ran at the enterprise edge. Right, so a small version of what the macro network looks like. And that allows us to support smart manufacturing energy and retail use cases, things like that. So in this category of partnerships with people like ABB, Schneider, Omron Dell, now to actually enable people to solve a different problem, not just a connectivity problem, but the efficiency, the product quality, the yields in these different sectors, manufacturing, energy, utilities, etc, right. And so what many of our customers in these industries are actually doing is using data analytics and AI to process data at the edge to accelerate their business decision making? Right, and we've got examples of automakers that are developing the next generation of electric autonomous vehicles, logistics and shipping ports, next generation computer electronic factories, safety applications in energy utility, you know, got to make sure that now if a fire breaks out on an oil rig, I got to know about it quickly, so I need to process that and deal with it. In the retail space, we got customers that are using private wireless or random access with video analytics, and that allows them to provide contactless customer experiences improve the efficiency of the point of sale. But this also leads to increased revenue as well as reduction in fraud and losses. And so just as we, as I mentioned before, we just came out of mobile Congress. And we made a number of these types of announcements with our partners leading up to and during mobile Congress for these types of cloud run and edge use cases.
Jo Peterson:That sounds pretty cool. I will. Now, you know, this might feel a little interesting, but I'm going to read you back a quote that you said, you were quoted as saying cloud ran will see a stronger emphasis in 2023, along with a desire to refine multi cloud partnerships, and edge strategies. How do you see the refining of multi cloud partnerships, and that strategy is playing out in the enterprise customer space.
Ian Hood:So we kind of talked about where the cloud ran fits in, and how that drives to potentially private wireless enterprise edge. And this enterprise edge location is really a key target for systems integrators, operators, and cloud providers to offer these types of secure innovative edge services across all these verticals, right. But what we've also got going on is that many of our customers have already been using our platform, in combination with a fully managed version of OpenShift. On top of the public clouds, to develop and deploy their applications, you know, auto manufacturers, banks, they develop their applications and deploy them today in hybrid cloud, using OpenShift as a platform. And so now what we find is that, as we look at that edge location, we find our platform being extended out to those Edge locations, and the customer will look at multi cloud partnerships to enable these application deployments. And they will take advantage of that same OpenShift platform on things like AWS, I posed wavelengths as our edge stack. And really what the customers are trying to do is to balance how they utilize the hybrid cloud. And this balance is going to be based on sort of data security, sovereignty issues, as well as locality and whether or not these application ecosystems, you know, how large and capable Are they from the various public cloud providers, so kind of a interaction between multiple players to create what was a hybrid cloud between enterprise on one cloud provider to multiple cloud plays, depending on where the data is, and what my data security requirements are. And so all these edge use cases we're talking about are going to require the partnerships across those systems integrators, handy operators, and this will enable them to actually get that business value from the data at the edge, whether it's on premises or on one of these edge clouds from a hyper scalar.
Jo Peterson:from an architectural standpoint, it feels like you've just sort of up leveled and multi cloud, you know, we were kind of in this multi cloud thinking. But this sort of extends a little bit further,
Ian Hood:it does, it definitely takes it further, because what is new is taking that multi cloud to the edge, right in saying, I want to use multiple hybrid cloud providers. But you know, AWS may be available in this location in this country. But I might need to use Google somewhere else, right. And so that's kind of where the, where it gets a broader collection of multi cloud endpoints, it's just a matter whether it's the big public clouds, or these Edge locations, or my own on premise variations. Right. So that's how it becomes a, you know, a broader collection of things that we're going to have to figure out how to interconnect those. And that's another key thing that we're spending a lot of time on now is how to actually interconnect between these different edge and large public clouds.
Jo Peterson:It feels almost super cloud ish, it feels like we're talking about an abstraction layer on top of public cloud, does it is it, it's got some of that feeling for me.
Ian Hood:We're not quite there yet. But actually, we're working on something of that nature. So what goes on right now is that we basically build an on premise to a public cloud, and we, you know, scale back and forth between those unbalanced workloads between the on premise and the public cloud. And in most cases, you find application a with public cloud provider, a, and then application B with a cloud provider be based on kind of those relationships, there's not really a kind of a, you know, a talking to be on premise. And on public cloud as kind of where it goes, multi cloud is still kind of, you know, in distinct pairs. And so we are going to start to see a need. And we've seen an application for this. If we take, I want to actually do AI. And I want to keep my data local. So I'm going to do that on premise. But I'm going to go to AWS and I got to do my training over there. I gotta go to Google. I'm going to do my data analytics over there. Now I need this application to interact with all three of these together. Right? And so we at Red Hat have a technology that allows us to actually do this and build a layer seven connection over other top between the on premise and these public clouds that secure that scales that allows us to temporarily build these connections just for the amount of time that I need to have access to that application and make somebody think the application is running somewhere else, when it's actually still local on premise. So this is an area that, you know that abstraction above the clouds is probably the next sort of the next thing we're gonna start to see in the financial sector and telco and large industry.
Jo Peterson:Oh, that's cool. This, it feels to me like almost a software defined approach.
Ian Hood:It really is a software defined approach. And the name, we call it at Red Hat, we call it application interconnect, which is kind of a variation from API interconnects. I make it a I just call it an application cloud instead of an edge cloud, because the applications are talking to each other, back and forth. And it's the same approach. If you think about it, in the way that whatsapp on your phone communicates between an iPhone and an Android. It's a connection over the top that's secure. So your two phones can talk to each other and not know that there's anything different between them. Right? This is doing this for business applications, instead of just voice communications and texting and things like that.
Jo Peterson:Oh, that's pretty cool. to have you back to talk more about this,
Ian Hood:you can maybe come back and talk about that one. As I say, we've been around a long time. And we've been trialing with a number of customers. We're getting some traction with it. And they say, you know, we should be seeing it coming too late in sort of production J fairly soon.
Jo Peterson:That's cool. Thanks for sharing.
Maribel Lopez:Yeah, we're really seeing a lot of activity in sort of this, I think to the point you're both getting at around super close, like how do we get to that next iteration, where we can leverage all the different benefits of everything that's available to us. without it feeling like we have to manage 15 separate things all the time, data back and forth all the time, because nobody wants to copy their data back. And
Ian Hood:that's the reason we have looked at this particular thing is not having to copy the data and multiple copies, and sort of the vision of multiple data lakes, you actually might be able to kind of eliminate some of those. The other thing is that, since we're doing an over the top, whatever's underneath from a network perspective, we don't care, right? So you'd have to create a whole pile of IPsec, VPNs, pinholes and firewalls, all kinds of complex things underneath to solve this problem. And much of the industry is actually attempting to solve it from the network layer. Right doing it that way. There's lots of technology out there in that in that category. We're trying to solve that the application layer for the application developers to extend their applications across these different environments.
Maribel Lopez:Makes total sense. Well, I'm excited to hear more about that in the future. But we have entered that time, the podcast, where I'm gonna ask you if you have a fun fact to share with the audience.
Ian Hood:Sure. So I got a fun fact. Is that, so I live in Washington State in Bellingham. And there's a part of Canada that many people know called Victoria, British Columbia on the island. And it's actually south of where I live.
Maribel Lopez:So you're in the US, but you're somehow north of Canada.
Ian Hood:Right. So think about that one for a second. And there's another fun fact for the same place is that there's another part of Washington, known as point Roberts, that if I want to get there from where I live, I need to cross two borders, wanting to get into Canada and want to get back into the US to get the point Roberts. So I live this really crazy part of the Northwest, where we have, you know, Canada to the south, and this other little chunk of United States that's hanging off the edge of Canada by itself.
Maribel Lopez:See, I like to believe that this proves that anything's possible. And that you just have to be open to the whole thing. Well, Ian, thank you so much for your time. We had a great time chatting with you and looking forward to seeing you back on elevate the edge in the future.
Ian Hood:Great. Well, thanks so much for umbrella. Joe. Appreciate you taking the time. Have a good day. Thank you again.
