5G Factor: New Cloud, Edge & API Moves, Episode 63

Hello and welcome everyone to the 5G Factor. I'm Ron Westfall, research director here at the Futurum Group, and I'm joined here today by my distinguished colleague, Tom Hollingsworth, the networking nerd and event lead at Tech Field Day here at the FU group. And today we will naturally be focusing on the major 5G ecosystem developments that have caught our eye, and that includes, uh, for example, some new developments in the cloud area, as well as an update really on network APIs.

And with that, Tom, welcome back. It's great to see you. This is an effect, like a trilogy installment for us.

It is, and I've been really busy since the last time I was on, so, uh, we're we're really getting the mobility Field Day event tied together. We just announced that we're gonna be hearing from, uh, companies like Ubiquiti and a new one Eno, who's doing, uh, some AI based planning in the space. So it's, it's gonna be exciting times, but luckily I was able to break away for a few minutes and, uh, jump on here with you so we can talk about some of the exciting things going on, on your side of the fence.

Well, hey, it's fully appreciated and yes, the Twain R meeting. I'm certainly excited about the upcoming mobility Tech Field day and thanks for that update. Uh, so this is definitely a scoop, uh, for the folks out there.

And as a reminder, this is taking place May 15th through 16th. And yes, we have an all star lineup, including the new additions that Tom alluded to. And with that, let's basically, you know, continue on the same theme mobility ecosystem and what is going on.

And we're looking at basically the challenge of deploying and administering private 5G networks as well as edge computing. I think we all understand this is not a simple task, and as such, it typically requires high levels of expertise, security, and flexibility. Plus it has to integrate with the existing IT systems and processes.

So that is a whole lot just to address and to actually, you know, meet these challenges. Microsoft developed Azure private MEC, which is a solution designed to streamline and accelerate the deployment of private 5G networks on the Azure Cloud, uh, platform. Now, Azure private MEC or MEC can provide a fully managed platform that can allow, for example, manufacturers to deploy and operate their own private 5G networks across the mobile edge computing realm.

And that includes full control over their data and applications. So I think that's going to be increasingly important, is giving them that peace of mind, that confidence that their data will just remain, that their data, and that by using a private 5G implementation with whatever else is already deployed, including wifi, for example, there won't be any compromises when it comes to security. Now with that set up, let's delve into two examples of where we're seeing Azure Private MEC and its ecosystem partners helping manufacturers specifically to use 5G for basically advancing their business outcomes.

Now one is Pegatron, which is an electronics manufacturer that produces products such as laptops, smartphones, and gaming consoles. Something that's fun to all of us here. And as such, uh, Pegatron is using Azure private mech to accelerate its digital transformation and enhance basically its competitiveness.

And how's that? Well, by deploying the Azure private mech, I find that Pegatron is creating its own private 5G network that covers its entire factory, thus enabling high speed and low latency connectivity for its own devices and machines. Now this is enabling, uh, Pegatron to use edge computing to run data intensive and time sensitive applications on the factory floor itself, such as quality inspection, predictive maintenance and inventory management.

Things we've been hearing about for a while, but now are actually becoming real. And in addition to that is Wix, which is equipping front it's frontline workers with wearable technologies and solutions to improve performance. And how do they're doing that?

Well, they're using what they call the Walt Smart Radio, which is being used by every frontline worker that has, uh, complete access to each other as a result, as well as their office teams and basically the community of workers that they have to work with across other parts of the globe. Now, the company's Internet of Workers platform helps to reduce downtime, expand enterprise wide collaboration, naturally improve efficiency, as well as safety measures that can basically just make that a possibility real, that is have a workforce that is improving its productivity, but also the overall experience. IE happier workers now by using Azure Private mec, we VIX is providing low latency and high reliability communication that is well, quite simply scalable but also secure when it comes to that edge data processing and storage.

And as a result, I see that Azure private MEC also allows internet workers to integrate with other Azure services. So it's not just this one specific solution, but it has to clearly interwork with other capabilities such as Azure, iot Hub, Azure Machine Learning, Azure Cognitive Services, as well as the existing IT and OT implementations that are already in place. And so with that, Tom, you know, what are your key takeaways from, you know, the Azure private ME proposition as well as, you know, what's going on with overall private mobile edge computing, uh, trends?

I like where Microsoft is headed with this because one of the things that a lot of people have to understand about what we would consider to be the enterprise is that it's not always an office. Um, I talked to a great company this week, uh, that uses the term uncared enterprise to de to, to kind of describe what they're doing. And when you think about a manufacturing floor, that is the perfect example of what you know would be considered, uh, you know, mission critical kind of, um, application to a business, but yet there's it, it doesn't have pretty LED lights.

It, it doesn't have these beautiful carpet fibers. In fact, usually everything's covered in a thin layer of dust, right? It's also a very hostile RF environment.

And one of the things that you need to understand about that is this is the perfect deployment scenario for something like a private 5G radio network because it gives you wide coverage, but it also makes it so that you can deploy these boxes to these large facilities. And one of the things that a lot of people don't understand is like these milling machines and these operational machines, a lot of 'em still run on some very old technology. I've heard all the way back to something like arc net, because once that machine has been certified to operate, it can't be adjusted like it has to be re-certified in order to be adjusted after the fact.

So by putting these devices kind of in front of those as it were, not only do you provide connectivity back to a cloud and to other cloud services that you can offer, but you also gain the ability to run things on that edge compute cluster, which allows you to do things like collecting analytics and being able to provide up to the minute kinds of feedback on things. I mean, how many times have we heard, uh, the amount of processing that these systems can do, you know, hundreds of units per minute, you know, but if you needed to say, okay, well how many of these units, uh, rejected bad work in the last three hours? Well, we don't know.

Uh, we just know that it rejects bad work and we run those stats at the end of the day by having some kind of an edge compute device in the middle, you can then, you know, batch those statistics and upload them up to some kind of a cloud application that can give you up to the second information, you know, hey, we are seeing, uh, a higher than normal amount of, uh, bad work being processed from this particular line. We need to go investigate what's going on so that we're not costing ourselves a lot of extra goods in processing things that aren't going to be able to work. And Microsoft is hitting on this key that all of these enterprises are connected now, these are no longer islands of, you know, machines that work and it takes a lot of glue to tie them together above that level.

If you can integrate these edge computing devices and pull those pieces together at the, the where they interface into the network network, you can leverage a lot of services that people are very familiar with. Yeah. And I think it's, uh, clearly a positive that they have two real world customers who are implementing this.

So, you know, we're getting past, you know, what about conversations to, okay, you know, what are some of the specific challenges that are going to have to be fulfilled in order to, you know, make this work on a optimal level? And I like the arc debt example, if I'm not mistaken, they're still casinos that use it for their one arm bands for the slot machines. So I think that's a, a wonderful way of looking at, yes, this new technology has to interoperate and interwork with existing implementations in order for that enterprise to really sign on and then start, you know, thinking about, okay, how can I evolve the overall implementation to take advantage of new capabilities such as predictive maintenance?

That, uh, you touched on Tom, and I think this is all, you know, really demonstrating that the 5G ecosystem is maturing that one of the key barriers that we, uh, have seen is that it was just the availability of, you know, for example, 5G uh, technology as well as mobile edge computing capabilities that wa was really a barrier for many of the large enterprises. And this is a demonstration that, okay, that will increasingly become less of an issue and it will become more of the practical implementation, uh, items that will be getting more top priority. And so thanks for those excellent insights.

And now let's turn to recent moves by another we'll partner yet competitor of, uh, Microsoft Azure, and that's VMware or better known as VMware by Broadcom, uh, these days as they successfully completed, uh, that acquisition. 0 came out, uh, recently, and I'm seeing it raising the competitive stakes really across what can be characterized as the telco cloud realm. And why is that?

Well, this release is solidify numerous capabilities into a unified platform. So this is gonna be a common theme here that can potentially give telcos the flexibility to modernize their network according to their business objectives, as well as tap into those automation and insurance capabilities that are well basically essential. And as we see it, the rigors of designing and managing services that combine the complexity of coordinating virtual network functions or vfs, uh, along with cloud native network functions or CNFs, really demands a horizontal infrastructure approach that supports basically the deploying, the running and managing of both the VFS and the CNFs with consistency, efficiency, and automation.

Now to streamline the process of adoption and modernization, VMware by Broadcom is simplifying how customers can implement and deploy the VMware Telco products themselves. 0 of VMware Telco cloud platform that has two packages, that is the VMware telco cloud platform essentials, as well as the VMware telco cloud platform advanced package. So it's pretty intuitive how they work.

And as a quick summary, the VMware Telco Cloud platform Essentials gives the decision makers a really, uh, a way to implement this and a fast, easy approach to deploy the infrastructure and manage the solution. It's basically kind of a run and gun type of proposition, but we know it's gonna vary and we know it's going to require some support in some scenarios. But the good news here is that VMware is giving this option out there.

Now, if you have more advanced needs, well ta-da, there's the VMware Telco cloud platform advance package that includes the automation capabilities and the essentials package, but also adds things like network slicing, multi-cloud support and network function automations such as the VFS and the CNFs that are already touched on. And this, I believe, can furnish a competitive automation and orchestration solution for core telecommunications network, something that is clearly in demand and is certainly being, I think, elevated in terms of its channel possibilities. 0 and its implications for the industry, certainly for the rest of this year?

You know, Ron, Ron reminds me of when Avaya was trying to jump on the bandwagon of, of getting as many of these things of their old phone systems, their North Star systems that they'd acquired from Nortel to be SIP enabled. Well, those devices are never gonna be able to be upgraded. They're never gonna be able to work effectively with that technology.

So what they did is they created a middle layer, they created a server that included all of the advanced functions that you could possibly hope for to be able to interface with that old equipment because it was a whole lot cheaper to sell the middle box than it was to convince a company to rip out their existing infrastructure and replace it. And I see VMware by Broadcom doing that exact same thing here with, uh, you know, this technology because a lot of these companies, they're looking at what it might take to get to more of an upgraded solution in the future, but for now, they don't have that capability. You know, telling a a organization they have to go and rip out a whole bunch of hardware is a pretty quick way to convince them not to buy anything from you.

But by creating this telco cloud platform, what they're effectively saying is, we can help you by building something that allows you to interface with that old equipment, but to be able to automate it, to be able to orchestrate it, to be able to manage it more effectively, and to get a little bit more life out of it while you're trying to gain more functionality into it. And when you familiarize yourself with the interface that VMware uses, well then when it's time to pull some of that equipment out and replace it, why don't you just go with the thing that you, you already know already. So I see this as kind of a, a double edged or a double pronged attack if you'll, they're creating an a, a middleware glue layer if you'll to be able to squeeze a little bit more life out of this equipment, but they're also creating some kind of an, an interface so that the users who are familiar with it will be more likely to go with that solution in the future.

And I think that this is a valuable tool from VMware, especially now that they're owned by Broadcom, because by making it sticky, Broadcom can continue to get that recurring revenue from their subscription based models, and that's really gonna make their investors happy. Yeah, and I think it'll also make, uh, customers happy, channel partners happy. And yes, there's been some criticism out there, but I think, uh, when it comes to this certainly, uh, area, this is going to prove, uh, just that beneficial for all parties.

And I, I like the emphasis on, you know, not propositioning a telco with, you gotta rip out hardware in order to make the solution work. This is reinforcing, you know, uh, the theme that we touched on at the beginning that is these solutions have to inter work and interoperate with the existing implementations. And I think this is definitely a way that VMware is really meeting that fundamental proposition.

Yes, there are greenfield exceptions, there might be some other exceptions, but that's that, you know, those are the exceptions. The vast majority of the telecom market at least is going to be how are you gonna fit into my implementation? 0 is so welcome and is actually meeting those, uh, types of demands.

And with that, wait, there's more. 0 that disaggregated ran functions are going to be required to really in sta instantiate on a ho horizontal platform that can be deployed at the locations that best serve their functional purposes. So again, it's giving the operator that deployment flexibility, that implementation agility, and by also including cloud native network functions, or CNFs as we, uh, define them, can be managed efficiently at scale and need to be, you know, quite simply manipulated to turn business objectives into those productive outcomes.

Uh, we've been talking about. Now, I discern that a horizontal platform can make disaggregation an operational reality. Why?

Well, the use of a horizontally consistent architecture that supports cloud data principles can cultivate the construction of logical end-to-end networks that quite simply are going to meet different 5G service demands. And to that end, VMware telco cloud platform ran, can deliver a horizontal platform optimized for running virtualized baseband functions or virtualized distributed units or vdu, which is a common, uh, term in, you know, the telecom industry as well as the virtualized central units or VCUs from various vendors. So again, the multi-vendor aspect here is basically table sticks while also naturally meeting, you know, the usual stringent performance and latency requirements of, you know, radio access networks.

And so from your view, Tom, how is, you know, VMware, you know, further augmenting its overall telco cloud proposition? What about it that's jumping out and, you know, why disaggregated ran is needed to really, you know, make a difference when it comes to telco operational flexibility? So everything I said about their, their, uh, platform in the previous story applies here, but there's that extra added wrinkle when you throw the RAN thing into it because as we've learned over the years, you know, ran base stations have completely different functionality based on where they're located.

A ran base station in a city is gonna look very different from a user pro profile perspective, from an application perspective than a rural ran base station. You know, in, in a say some kind of a metropolitan area, you can have distributed functions in different areas. And if you get, like, you know, say for example, during a sporting event when people will tend to aggregate in one side of this city, you know, you can kind of distribute that load a little bit, but in a rural area, that thing has to do a lot more heavy lifting because the users on that base station are so, you know, geographically dispersed.

And I think what VMware is effectively saying is, what we're trying to do is build a platform that allows you to add pieces to those, um, puzzles, if you will. And, you know, in a, in a more dense environment, we can use these platforms to augment each other when we need to distribute load, when we need to add application support. But when you get to an area where you have a much more distributed user base, and you and your units seem to be a little bit beefier in order to handle, you know, the fact that that tower might be serving, uh, you know, thousands of subscribers without any kind of, you know, real reliable backup, um, by making that a horizontally scale scalable platform, you're getting the consistency of what they offer, but also the capability to build as you go so that you don't have to worry about it.

And I remember this from my old, uh, days working in, in a reseller environment, you, you open that closet door to that base station virtually, if you will, you never know what you're gonna find in there. Oh, great, we can't do the thing we were planning because this equipment's too old to handle that. As these devices become more commoditized, as they become more reliant on kind of commercial off the shelf hardware, the layers of software above that are the things that are doing the heavy lifting in order to really help people understand how these things are supposed work.

And again, VMware is kind of building this out. They're providing the i, uh, customer demand is effectively, and by creating that and by making it easy for people to use, what they're doing is, is they're making people wanna stick with VMware by Broadcom, which I mean ultimately is every company's goal, right? We want you to continue to use our stuff, but they're also building a platform that allows to build capability in the future.

Because that's the other thing that companies really wanna avoid. They don't wanna buy a one and done kind of solution for this. You know, we're gonna stick this in for a couple of years, it's gonna get us to the next, uh, revision, but then we have to throw it all away.

VMware's messaging on this is, is much different. It's saying, you know, right now you need this interface layer, but in the future, that interface layer can be transformed into a much more robust software solution, and that's going to carry you into the future so that you don't have to rip us out when you're getting rid of the equipment, when it finally reaches its end of life day. Yeah, those are excellent points.

And yes, it is about that deployment versatility. We all know, like small cell densification, for example, is gonna be much different in an urban, uh, environment than, you know, suburban or rural implementations. And it just doesn't end there.

There's the antennas, there's the spectrum allocation, there's just so many different factors that feed into, okay, how can we optimize really the RAN implementation to, you know, meet all of these different technical and business objectives. And there might be a bit of good news here on the horizon and how this can be helped regardless of, you know, the RAN implementation, whether it's a traditional RAN implementation, a virtual ran implement implementation, a cloud ra, uh, ran implementation that is, you know, using open ran principles as, um, as an example. And that is, you know, network APIs, which has been a, a very popular topic in Q1, including naturally mobile, mobile Congress 2024 after our friend ai.

So I had to get arm roll mention to ai, uh, before, uh, we could conclude our conversation. Uh, but the reason is is that Vonage, which is now part of Ericsson, uh, for a while now, is partnering with, uh, Stel to integrate Vonage API platform with singtel's orchestration platform, which is called Paragon, to boost these innovation opportunities for enterprises as well as the telcos to access a global ecosystem of communications network and edge APIs. So it's not just about network APIs per se, but also edge and communications APIs.

Sometimes they overlap, but I think there is warrant for, you know, demarcating these different types of APIs because they're all gonna make a difference with what. Well, let's look at Paragon itself. Now.

Paragon enabled telcos are, include Singapore, naturally Thailand, Spain, Taiwan, and Indonesia. And as such, they're going to be looking to integrate the Vonage and Paragon APIs into a, a unified Telco and edge API library. Now this leads into allowing, you know, software vendors, enterprises, and developers, and this is, uh, the part that I'm emphasizing to look at and use the APIs to not only deliver digital services that better speed and scale over, you know, what are basically hybrid networks now, but also the hybrid edge and public cloud infrastructure across these different geos across these different verticals, uh, basically across any market, uh, that is being targeted.

Now, uh, to uh, preface this, I, uh, talked about how can this help with the RAN implementations? Well, certainly one, I think use case that has gotten a fair amount of emphasis is the energy efficiency aspect, but through network APIs, there's potential for the telcos and whoever else is, uh, being impacted to figure out how can we do that. Again, distribution of energy to make sure that the, uh, telco itself is certainly meeting its, you know, sustainability goals, its ESG objectives and so forth, but also to, you know, just that lower the bill, you know, but without cutting off the service for like E 9 1 1 and other vital services.

So it's a balancing act, but this is where third party developer innovation can play an increasingly important role. Now, this is not new to the industry network APIs. Uh, after all, we've seen, uh, similar initiatives in the past that just rest and soap that sought, you know, broader API adoption in the telco room.

However, those results were decidedly mixed, uh, or perhaps, you know, somewhat modest. But with that, uh, Tom, uh, how are you seeing this develop? How are you seeing, you know, network APIs coming on board and you know, this time it's different perhaps now what's going to be different and, you know, what are the prospects here these days When you look at the way that people consume communication services?

Now, this is not ma Bell's good old fashioned pick up the phone and get a dial tone. They are looking for advanced services. I mean, when's the last time you went out and bought an answering machine?

The answer's probably you haven't. I mean, even if you have a dedicated landline service, you probably have voicemail that's being offered by one of the, uh, the, either the company that you're with or by some other kind of partner company. Um, 9 1 1 is a perfectly good example too.

I mean, it is a requirement in the US that you have to provide 9 1 1 services to people. Well, how are you doing that exactly, and how does an upgraded phone system do that? I spent a lot of my time working with telcos trying to figure this out back in my, my reseller days.

And it's, it's a difficult proposition to crack whenever you're dealing with modern, uh, public safety answering points and things like that, unless you have some kind of an API interface into that. So by creating an API interface down at the user layer, something that Vonage can do, you're opening up the world to all of these kinds of, um, you know, services that you can resell. Uh, voicemail's obviously the biggest one, right?

Sure. But what about call filtering? I mean, Lord knows that people who still have landlines are probably the ones who get the majority of the, uh, you know, uh, spam callers or the, you know, the kind of, uh, scams or, or even things that you may not want, like political, uh, uh, surveys and things like that.

So as a service, as someone who maybe wants to talk to Vonage or, you know, even somebody who's leveraging them like Singtel or any of these other telcos, I can offer that service for a small monthly fee and you know, oh, well, yeah, I'll, I'll subscribe to this and then I won't get these calls and things like that again, only possible because of the AI API interface. 'cause if you look back to the way that these things had to be implemented before, they don't play well when there's no way for these systems to talk to each other. And so I think what you're gonna see is these VoIP companies who are kind of, you know, continuing to, uh, drive innovation, are gonna continue to offer these API integration solutions so that developers on the cutting edge of that technology can provide the kinds of services that pe make, people wanna keep their landline because that's one of the things that we've seen a lot over the years yet here recently, is people are just shutting them off.

I mean, they don't need them. They carry a mobile device almost all of the time. Why would I even bother?

And it's actually causing a lot of companies to look at whether or not they even wanna keep those services or whether they need to divest them into some other kind of vehicle so that it doesn't affect their bottom line. Because as you know, nice as it is to be able to have that recurring revenue. If you can't spruce it up and make people wanna buy it, then it becomes kind of a, you know, a yoke around your neck that you're gonna have to carry because Ken, you have, if you, they have a landline, there are certain services that have to be mandated for it.

So I think that this is a way to kind of offer a way for telcos to provide the kind of services that are gonna wanna make people continue to use those landlines and augment their existing mobile device, uh, footprint. Yeah, and I think those are all solid points, Tom. And I think one thing that also has to happen, and Vonage is putting priority in this, and this came outta conversations that Mobile Global Congress, is that it's making the network API experience itself more friendly Now that is, you know, basically reducing, you know, some of the overhead that has come up with just trying to use the API that you know, is, uh, going to get these developers incentivize to want to, you know, work with this network to want to, you know, provide some, uh, new capabilities.

But it's like, uh, how many people have their business identification codes available when they're trying to sign on? It's not readily available in many cases, and that could be a hindrance. And so it's something as basic as that that can be, uh, a difference maker.

It's about the blocking and tackling of making the network API experience itself, just that more developer friendly, more third party friendly. And I think that's going to help, you know, answer the question, what's different this time? And there's other factors in play.

You know, naturally 5G standalone will become increasingly prevalent during the course of 2024. And that I think will also make, you know, the 5G core capabilities, all that, you know, more programmable and accessible and so forth. And so this is, you know, coming together to, you know, make, uh, again, 5G services more attractive potentially, and again, uh, allow the telcos to really monetize on a broader scale and on a more diverse scale.

Okay. And so with that positive note, thank you again, Tom, for joining, uh, the 5G Factor. Certainly looking forward to our next 5G Factor and of course mobility tech Field day, you know, a few weeks from now.

Well, thanks Ron, and I'm really looking forward to it as well. And uh, I, we're gonna have a great time. You'll be there.

com on the 15th and 16th of May. Thumbs up. Uh, exactly.

I, I know I literally will be. And with that, thank you to our viewing audience for joining us. Uh, again, we appreciate taking the time out and don't forget to bookmark us naturally, uh, in terms of, you know, remembering 5G Factor.

It's basically a once a week type of proposition. And with that, thank you everyone and have a happy 5G Cloud and network API day.