In The Arena by TechArena - Exploring the Intelligent Middle Mile with Ribbon Communications

Episode Date: March 12, 2024

TechArena host Allyson Klein chats with Ribbon Communications’ Jonathan Homa and David Stokes about the progress of 5G proliferation and the importance of the adoption of the intelligent middle mile in reaping full benefit from 5G services.

Welcome to the Tech Arena, featuring authentic discussions between tech's leading innovators and our host, Alison Klein. Now, let's step into the arena. Welcome to the Tech Arena. My name's Alison Klein. We're coming to you from Mobile World Congress in Barcelona, and I'm so delighted to be joined by Jonathan Homa, Senior Director of IP Optical Solutions Marketing and David Stokes, Head of IP Solutions and IP Portfolio Marketing. Welcome to the show, guys. Welcome, welcome. It's great to be here. It's such an exciting show this year. I know. It's incredible, isn't it? And it just feels vibrant and alive in a way that it hasn't since the pandemic. So it's really wonderful to be with you guys today face-to-face. Why don't we just start with introductions? Jonathan, do you want to go first and just introduce yourself?

Sure. Jonathan Zlomat. I'm responsible as part of our IP optical solutions, focusing on really two aspects. One is the optical networking aspect, but of course, how this integrates with our IP solutions and also providing expertise and how we pitch our solutions to the marketplace around the SDN control aspects of our IP optical solutions. Fantastic. And David? Yes, welcome. And David Stokes, I'm responsible for our IP routing portfolio and how we can use that to build solutions to the end users. That's my former company role. I have a massive

passion around the fact that telecoms can be used to bridge the digital divide, enable new services that benefit society. We're 16-year-old twins. I want a future for them. And I think it's really important that telecoms have the chance to do that. That's a bold and ambitious goal. And I love it. We're at Mobile World Congress and everyone's talking about

proliferation of 5G. And we talked to you guys last year about the proliferation of 5G. What do you think is different in the conversation this year? And where would you say we are with adoption of 5G across all of the various use cases that it was targeted for? I think we've reached a point in 5G where the cap expense is peaked. It may be declining, it may not, but it's certainly peaked. And we've also, at the same time, if you know the data hype cycle, we've got the height of unexpected expectations at the peak, and everyone's thinking they can do everything.

I think we're right now on 5G in the user community, now right down in the trough of disillusionment. 5G's out there, nobody's getting it. Those who are getting it, it's not really delivering what they expected. Sales providers are struggling to make money out of it. So we're right there in the trough of disillusionment. And that's because I believe 5G has gone out there just standalone, and now we need to have a backhaul backbone,

a middle mile network, which is aimed to connect this access rollout to services and applications, which A, can make money for the end users and the mobile network operators, and they are mindset sobox, which can start doing

all those societal beneficial features that have been promised in 5G. Now, you guys have been talking about this concept of an intelligent middle mile. So I've got to ask the question, what is the middle mile and why is it so intelligent? Well, I think we should start off, Alison, with talking about what is motivating the middle mile and why it's rising so much in the conversation in the industry today. So you mentioned 5G, and of course, there's been massive investment in the last few years in 5G, but it's not just been 5G. It's been equally massive investment, often funded by governments themselves within the fiber access. So really what we're talking about before we get

to the middle mile is we have to talk about the last mile and we have to talk about the first mile. So the last mile in the common talk of the industry is RV access technologies, the wireless access technologies, 4G, LTE, 5G, people are already talking about the next gen on top of that, but as well as the fiber access. And what this is doing is opening up the floodgates of broadband capacity to the end users. It's making the potential in terms of having a very rich variety of services. And these are going to connect, of course, to the first model, what's also called the core of the network, where the data centers reside, where the applications reside,

where the intelligence, where the services reside. So what you now have is an issue where you have all this pent-up services capacity in the last model, which you really need to find a way to properly monetize and be able to transport that efficiently and in a way to maximize the services value with the services sitting in the way to maximize the service's value with the services sitting in the data centers in the core which by the way doesn't necessarily have to be in the physical core more and more we're seeing this distributed along the geographic network so the middle mile well it's called the mile can actually span from tens of miles to low thousands of miles

in terms of geographic distance in reach with the overall network. And it's really providing the connectivity between that services potential being built up in the access and the actual services intelligence sitting in the core. And you want to find a way to transport the services efficiently and be able to maximize the services value. And we can talk a little bit about services of various,

and I think David will do that in terms of, say, offering differentiated services, but also doing this in a way that lowers the total cost of operation for the operators. So when we get into the intelligent middle mile, it's how we make this middle mile intelligent that we could pursue those two goals simultaneously. Service awareness and total cost of operation. Maybe, David, you want to talk a little bit more about the services aspect. Yeah, absolutely. And as Jonathan was talking about,

I don't know why, two songs came up into my mind. And you could have the way we have today where everything on that middle mile is nailed up and the operators and end users are handcuffed to what they can provide. And suddenly Chris Rears rode to hell. Just sprung up. I wasn't sure where you were going and it wasn't there

but keep going the reason they're being distributed is because it's all in the cloud and then immediately I'm thinking stairway to heaven so I think in song so on that side I was aware of it

by getting back to the CEO and I'm now thinking of the lady but anyway the promise of 5G and of the new services, we've even fixed Fiverr. All the services have different requirements on them. So you might have, you know, the classic one that people talk about is online cloud payments.

Right. And they pay a lot of money to guarantee latency and also availability. They're in a part of a game which is very latency dependent, you know, the multiplayer killing games. Ability

to draw your gun and fire the bullet till the laser or whatever it is slightly faster than the opposition meets you stay alive. Right. And again, now I'm thinking Ready Player One. So you need to have a network that can guarantee that low latency. If you're thinking e-health,

and everyone thinks e-health, science fiction, robotic surgery, I'm thinking of my mum. She wants to stay in her house for as long as possible but she's probably at a point where she needs assisted care that costs governments a lot of money and she's not she doesn't want to be there

if you can start putting wearables it's not that wearables are uncommon we all have smart watches now start putting wearables on people and you can put monitors in hospitals and people monitoring those. You can't allow people to live in their houses for a lot longer and feel safe and comfortable. But that requires a lot of, again, it's got to be high availability and it's people's health information. It's got to be secure. There are very different requirements to the latency

requirements. So you need a network that can understand what the services are and what they need to do and deliver the right parameters across the network so you can meet either latency or availability or security. And you're not sure you can do that or just

throw money and capacity at it. And then nobody can make any money. So you need one set of network resources to do that. And the routing layer is the first point. Through that, you can see what the needs of the services are and use your intelligent automation engine to

route that across. And then you go to the high-capacity services that Jonathan talked about. Well, I think just more just bouncing off your point a little bit, you know, the classic model for building sometimes. A middle-level network just might be throwing capacity at the solution. And then you're sort of going back into the basic internet paradigm of providing best effort transport for the services themselves.

And that works up to a certain point. But what happens is when you start getting into a lot of the services David's been talking about and really being able to maximize the value of being able to offer differentiated services to your customers, which enables the service providers also to make more money because they can sell these different services

with different tiers of pricing, that sort of best effort internet approach doesn't work anymore. Because in order to do that, first of all, you don't have any built-in hard service level availability guarantees. So you can guarantee service performance. And if you want to provide a better level of performance

for those more demanding services, to a large extent, you need to over-provision your network, which, as David said, you need at higher costs. So really what you need is the kind of flexibility in the network itself that you can do different levels of packet recognition, being able to do traffic prioritization, be able to do traffic engineering. A lot of the capabilities of today's IP networks enable you to do that. But then it's

providing the controls that enable you to manipulate the network, that you can actually match the services to these different types of engineered networks, so to speak, and being able to support this differentiated sets of traffic. Before I get into the sort of underlying optical layer, anything you want to add to that? No, I think that really has covered it. As Jonathan said, when you do that, it's not just about the network. You need to be able to have an automation engine that really understands that need and uses the right resources at the right time and has the flexibility to adapt over time.

As a new service or type is introduced, it suddenly might have a massive take-up in a certain user dynamic. And you need to have the network route that capacity that's coming from that access there to this new service, wherever that might be located. So again, as Jonathan mentioned, this nailed-up capacity

that used to be there with best effort needs to be far more dynamic within the network. We all know that mobile network operators have got a duty of care to reduce carbon footprint. The middle-of-the-world network itself doesn't embed the compute capabilities, but it has to be aware

of where those compute capabilities are located. And we're seeing what's happening today with a lot of the web-scale operators in order to provide better responsiveness for their services themselves, they're moving more towards a distribution of their data center capabilities, where they're moving, say,

micro data centers, what we used to call the mobile edge computing. It used to be something which everyone talked about. Now it's starting to be implemented so people don't talk about it anymore. But that's a fact. So what you're saying is it's there, and we're just like done.

It's not interesting anymore. And what the middle bar needs to be able to do is to be able to be aware of where those sort of edge computing capabilities are being located, and facilitate the flow of traffic for that efficiently, as opposed to, say, going to some central place

and then maybe moving back towards the edge or whatever which would be inefficient. So it's working together in terms of the addressing schemes, sourcing of the services where they're coming to, where they need to go to, and making sure the traffic flows efficiently throughout the network.

So yeah, to complement that, that edge compute and also some of the core functionality is absolutely part of the wider middle mile. And we really focus down on making the use of the transport to connect those resources is our goal. And our mission in life is to be best in class at doing that so that you can then use that compute in the most, for vendors who are experts in that.

And I think we all know who those are. So they can concentrate on their expertise, and we can actually deliver this stuff to them in a right way. Got it. Now, you talked about automation, which is something that's really important. Do you feel like services are being delivered by operators to take advantage of automation at this point, or do you think that we still have some work to do? When I think about automation, I'm thinking

about automation in terms of, from our perspective, in terms of running the middle model network. There's really two aspects to that. One is keeping it optimized continuously, providing a fixed set of resources, and then you can use automated routines to take a look at the traffic patterns, and then reallocate resources based on shifting traffic patterns even throughout the day. And this can be done both at the IP layer and also at the optical layer, which is becoming more and more configurable. There's also automation in terms of the services, maybe, which you were looking at. So here, when you have a service demand coming from customers.

So today, what happens is that service demand will come into the network operator. They'll get onto their various control systems. They will configure the service. They'll provision the service. So we make tools that make that particular process more and more automated, starting off, say, with services templates. They'll have to populate certain parameters within services templates, and then you can get into closed-loop provisioning cycles, whereby the network control, the SDN control

that I talked about, software-defined networking control, will automatically provision the resources, will actually download what's needed into the network, will actually test that the service operates, then it will activate it. Then from an automation point of view, it can continually monitor the service, and it's key that it's continually performing against the service level agreement guarantees.

One more aspect, which I think we'll be moving to, is at some stage, we'll be putting this into the hands of the end users themselves, that they'll actually be able to sort of have a portal into these control systems and be able to call up the services themselves directly and almost eliminating the sort of tumor operator aspect of this. Sure.

So again, I'm not thinking music again. I'm sorry. And I know that's my heavy role. Youth. I'm not thinking of heaven and hell. There's one theme here, I think. So Jonathan was talking about heaven and automation and everything automated.

Unfortunately, in the IP world, we still have people who think getting right down into the machine level stuff and doing everything against CLI is the way to do that. And when the world was best effort or when the network rooted itself, as that's what it does, that was fine. Now all these complex services and meeting SLAs, we need to help them step baby steps into the world of automation and the benefits those can bring.

I mean, for example, we've done studies that show using an automation engine compared to manually provisioning is between 95% and 98% quicker. Right. I mean, massively quicker. And you've eliminated all manual errors. Mm-hmm. Yeah. But for space on these guys who used to be right down there in the dirt, that they need to elevate what they're doing, is baby steps.

And it's really important to have the automation that allows you to doing is baby steps. And it's really important to have the automations that allows you to take these baby steps. I mean, when you talk AI to these guys, this is like science fiction. But it's reality. All the stuff has all the machine learning, the big data analytics. It's all running to do that.

Yeah. It's interesting. One of the things I was thinking about is that when you enter the trough of disillusionment, for your earlier comment, you often shift from a technology problem to a human problem. Yes. And I feel like our conversations at MWC are around, how are we going to get the human equation to utilize all of this new technology that's been created in 5G to get the full opportunity for the use cases that have been envisioned. We don't have time into getting into how we're

going to solve humanity's challenges today. So I will have one final question for you guys. You've talked about some of the solutions that Ribbon has delivered into the market, where you want to be in terms of the core capabilities that you're delivering. Where can folks find out more about the solutions and engage with your team? Well, first of all, here at MWC, they can come to our booth. I don't know if you remember the number. We're a hall team.

2C42, hall 2.1. Yes. And there we'll be able to show you our whole suite of solutions that we're bringing together in the intelligent little model. And by the way, just a final word, I don't know if you properly wrapped it up in terms of the intelligence. There's really two aspects to this. One part of the intelligence is how we perform the integration between the IP layers and the optical layers. We're not treating these as two separate entities. We're treating these as a unified whole. And it's really by providing this integration

from the planning aspect through the network commissioning aspect to the onboard service provisioning aspect, we're able to make maximum use of resources. And that's one layer of the intelligence. The other one, which we've talked about more, is the automation aspect in terms of the day-to-day running of the network itself.

So certainly come to our booth. You can learn a lot more about that over there. You can go to our website, ribboncommunications.com. Thank you very much. Or you can just reach out to David Stokes or myself, Jonathan Homa. You can find us on LinkedIn. Just Google our names, and I'll be glad to answer anyone's questions. Well, Jonathan and David, it was a real pleasure. I would love to have you back

on the program in the future to go deeper into this topic. Thanks so much for being here. Thanks for joining the Tech Arena. Subscribe and engage at our website, thetecharena.net. All content is copyright by the Tech Arena. Subscribe and engage at our website, thetecharena.net. All content is copyright by the Tech Arena.