Making the Open RAN Network Metaverse Ready

Sign up for Daily CommTech Brief here.

Session Abstract:

There are a number of buzzwords on the floor here in Barcelona, in 2022 and two of the ones that seem to permeate the event are Open RAN and Metaverse Ready Networks. Joining us to shed some light on why these are most prevalent are, Manish Singh - Head of Wireless Ecosystem Programs at Meta, formerly known as Facebook, Caroline Chan, VP and GM of the 5G Infrastructure Division at Intel and Mehran Hadipour, VP of Business Dev. and Tech Alliances at Robin.io.

Executive Speakers:

  • Manish Singh - Head of Wireless Ecosystem Programs, Meta

  • Caroline Chan - VP and GM, 5G Infrastructure Division, Network Platform Group, Intel

  • Mehran Hadipour - VP of Business Development and Tech Alliance, Robin.io

 

Transcription

MAKING THE OPEN RAN NETWORK METAVERSE READY

 

Abe Nejad: So there are a number of buzzwords here on the floor in Barcelona in 2022. And two of those seem to permeate the event. Those are open ran and metaverse ready networks. Joining us to shed some light on why these are most prevalent are Manish Singh, he's head of wireless ecosystem programs at Meta that is formally known as Facebook, by the way. Next to him is Caroline Chan, vice president and GM of the 5g infrastructure division that at Intel and last we have Mehran Hadipour,, he's vice president of business development and tech alliances at Robin.io. And welcome.



Caroline: Excited to be here in person.



Abe Nejad: We love zoom by the way, but not as much as in person. So good to have everybody in person again and also good to be or to see you for the first time in person at an event like this, where everybody's sort of under the same roof. So it's great to have you. So I'm going to start with you Manish, if you don't mind, let's talk. And I mentioned this, of course, at the top of the segment, let's talk about open ran, what's needed for open ran to sort of move forward. And is it ready for sort of prime time?



Manish: Yes, we are getting there. I think we already have a number of operators who have been doing active trials and deployments of open ran. The technology is stack is coming together. The specification work that's happening in Oran Alliance that's definitely made significant progress. And within TIP, in the telecom info project we've been pretty busy bringing the ecosystem together, bringing the operators, working with them to get the requirements in, working with the technology providers and vendors to bring the different parts of the solution stack together and doing a bunch of integration, work validation, working TIP community labs, and then supporting the trials that the operators are taking. 



And last year November, we actually published as part of the TIP blog, pretty detailed number of trials that are going. And currently, if you look at, it's around 40 plus trials across the globe that the operators are running. So yes, definitely tremendous progress, solid momentum, and right here in Barcelona, a number of operators already announcing their plan. So yes really, really exciting times for Robin.



Abe Nejad: Well, Caroline, certainly from Intel's perspective, open ran is ready for prime time, agreed?

 

Caroline: Oh yeah. I mean okay. So we've been working on this for like 10 years. And so we started doing the open ran, started with a virtualized, we're providing flex trend as a reference software stack. And more importantly, we really target a gear, our Silicon roadmap to be able to provide the right acceleration, including inline acceleration and 5g instructions set that we just announced. In fact we just announced ice lake with many operators supporting that. We also announce our next generation Silicon roadmap with many operators supporting that and I think what Manish said is true, we need to get the ecosystem robust. We need to do integration. We need to do testing, but look at the world today. We do have Rakuten that has launched a virtualized open ran network in the most dense network, Japan. We have Verizon New York City and we have Dish it's about to launch. And then we also have [03:41 inaudible] just announced. UK got to do this. So I really think that we are getting very close from the trial to actual primetime deployment.



Abe Nejad: Mehran from your perspective?



Mehran: I think open ran is reasonably ready for product, given the fact that we actually have over 3 million subscribers running 5g open ran in production is a good proof point, is not an exercise any longer. Some of the work that TIP is doing is quite interesting actually, because you're building different competencies and packaging of a solution. On the Roma project, we are building a wrap for TIP that integrates ran and core functions, and try to demonstrate on top of inter technology and show that we can operate the entire stack on cloud-native infrastructure. I think you'll see within a year that there will be number of open ran projects in deployment stage globally across.



Abe Nejad: Manish, certainly a number of organizations working groups working on Oran or open ran. Is there overlap there? Is there a lot of overlap or is there more collaboration there?



Manish: Yeah, I mean, look, I mean, TIP is working on open ran, Oran alliance is working on open ran and that's great. I mean, we do need different organizations with different focus areas to continue to make advancements on this. Oran Alliance, different working groups have made tremendous progress specifically, I would say on these specifications front. So that's very, very important because for open interfaces you need to have open spec and that works are being done very well in Oran Alliance. TIPs focus has been very much to take those specs from Oran Alliance, plus 3GPP and others, and then really drive the productization and commercialization of open ran. 



And towards that there's been very good collaboration between TIP and Oran Alliance. We have a liaison established, which enables TIP to leverage these specifications work. And then to give you another example, like tests and validation, that's very important. Again, when you disaggregate you open the ecosystem, then you need to bring the pieces together, and that should not take a big integration exercise. So how do we solve for that? TIP community labs have been a great place where a lot of this work has been happening and also in collaboration with Oran Alliance to share with you. There have been joint plug fest that TIP and Oran Alliance have done together. So again, I think great collaboration, both the organizations focusing on different parts and both needed to really move the industry forward.



Abe Nejad: Overlap collaboration as far as open ran?



Caroline: Definitely collaboration. I could say that having served on the TIP of for quite a few years, we really see, we really come into this very comfortable place for a true organization, TIP really take on the tour, right. We integrate, we test, we validate versus Oran Alliance really start defining the specs. I think the two organizations really started coming really well together.



Abe Nejad: Mehran anything to add to that?



Mehran: I think the benefit I see in a number of these Alliance organization is that more than just setting standards and API is actually interchange of ideas and collaboration, infrastructure, readiness, being able to understand the deployment models. Those are all benefits we see as participating, both open ran alliance and TIP, and many other standards committees. So I think they're definitely beneficial. I don't think there's a lot of overlap. They have different missions very often, but we're glad to be part of them. I think they're adding value to the whole progress on ran.



Caroline: Look, I mean, open ran is meant to be cloud-native. So that is a lot of learnings that traditionally the telco lacks. So it takes somebody like TIP who has a lot of the cloud ingredients and heritage to really come in and somebody like Robin.io, that really come in and really facilitate this integration of the cloud and the telco network. And I think that is both the challenge and opportunity of open ran, is to bring that cloud nativeness into it. So the networks more than just connectivity, it does a lot more than. I mean, Robin does a lot of that in this.



Manish: Yeah. If I may just quickly add to Caroline's point, within TIP the sharing is one of the key principles that TIP enables and to give you examples, for example, water fund did a trial in Turkey. This was one of the first early open ran trials. And there were a lot of learnings out of that, which would've water fund very publicly shared with the TIP community. And that's the value of bringing those learnings back into the community, sharing with the community. And that really helps all the technology providers to understand where the problems are. And also for other operators to understand how things are maturing. Because again, taking that example, they even shared how the network was performing, what KPIs they were able to achieve. So it's really that enablement TIP as a platform provides to the community and driving that effort [09:11 inaudible].



Abe Nejad: Before I was going to shift actually to a whole different topic. But before I do, I wanted to ask Mehran a question. So moving towards open ran specifically for service providers and telcos and operators, it's a technology shift as Caroline mentioned, but it's certainly also a culture shift as she also alluded to as well. I wanted to get your perspective on that.



Mehran: That's a great, interesting observation.



Abe Nejad: Is it of equal importance by the way?



Mehran: Quite a bit. I mean, the scale of 5g and the deployment complexities added to the fact that now we are talking about open standards, X 86 platform that are enabled for a specific use case. And you are looking at very new deployment models. And if you have to have automation built into the culture and to the infrastructure and deployment models that are very different from what the past used to be like, you got a box from the top three, is already preconfigured. Somebody raise it up to the antenna and you were done. The model is different. Now we have X 86 platform hosting diversified and authorized stack, that is changing daily, orchestration, automation, life cycle management, all those things have to be automated. All those learnings of DevOps like [10:32 inaudible] applied to the telco deployment is brand new cultural shift.



The guys who used to do operations are hardware support people. Now you are having DevOps people managing your telco stack, brand new model, but this is actually part of the complexity of getting to a diversification of the stack and [10:52 inaudible] of the stack. But everybody's adopting it, it's no longer an excess size. You see significant devop focus in the way the telco functions are being deployed. Telco functions are no more than other complex applications. How you deploy them, how to manage their life cycle has to be built in the model you support them.



Caroline: Yeah. And those are become even more important, as we talk about private network, you're asking enterprises to deploy something like that. It needs to be cloud native. It needs to be easy, needs to be automated, even more so than proper network.



Mehran: They even have less resources. 



Caroline: Less, resources and automation, zero touch.



Mehran: Has to be there. 



Caroline: Has to build in.



Manish: One additional input on this. One of the things we do look at also is how do you make open ran, easy to consume for the operators? And so to your point, it's very important that we also start to understand what are some of the call them, the golden configs for the open ran. So that's one of the other work that happens in the tip community lab is bring the ecosystems together, develop the blueprints, validate them, badge them, and again, share that with the larger operator community. But to your point, that's very, very important that we start to do that integration work and develop these validated blueprints.



Mehran: The other aspect that has to change, previously when people talked about automation, they looked at service automation. Automation now has to extend to all the way to bare meta. Somebody plugs in a box with nothing on it. That point it has be installed, managed, and entire lifecycle operations, operating system, corner upgrade, bio changes. You can be touching that, if you think about touching, has somebody actually have to do something to the boxes, there's not enough people in the company to do that. So orchestration has to extend all the way to bare meta for this to work for the open ran to. So that's why meta to service orchestration is the new concept of orchestration. We are not talking about just service management here.



Abe Nejad: I want to switch gears to another buzzword that we're hearing on the floor, particularly at this event, the metaverse. Certainly there's a number of issues tangential to that term, but a metaverse ready network is one of those. Manesh I wanted to start with you. What does that mean? A lot of us know what the metaverse is. So I guess we can talk about that, but what is a metaverse ready network?



Manish: Metaverse ready networks, basically networks that can enable metaverse. So metaverse brings presence, brings very deep interactions, immersive interactions and a new way for communication. However, it also brings certain new requirements with them. And in particular I'll say lower latency, symmetrical bandwidth and more. And that is what we need to solve for as a industry, as a community in a collaborative way to enable metaverse.



Abe Nejad: So where does the conversation around the metaverse network or ready network evolve to? So let's say a year from now, what are we we talking about?



Manish: Oh, we're talking about a number of things, like to lower the latency. What does that mean? Is there more role for edge to play? Is there even role for breakouts to happen, local breakouts to happen? How do you measure quality of experience in metaverse? So we've done quality of experience and voice, video, et cetera. Going forward we need to define the framework for what quality of experience is going to be. And to give you an example TIP today is hosting a birds of feather along with Telephonica to really bring the industry together, to start these conversations. 



What does it mean to build these quality of experience frameworks? And the last thing I also just say is the networks need to become more application aware and the applications need to become more network aware because network conditions can change to much lower granularity. And to again, enable Metaverse at scale, these are things that would be required, and that's what Metaverse ready networks are. And for more details lastly I'll say, the meta connectivity team published a really detailed blog around this as part of Mobile World Congress and anyone who wants to know more I'll definitely point them to that blog. 



Abe Nejad: I think the application aware aspect going both ways is really interesting. We should probably talk more about that at a later session. Maybe this time next year, we'll have some rubber meets the road conversations about that. From Intel's perspective, what's your take on the metaverse or the metaverse ready network, anything to add?



Caroline: We have had a discussion with Manish and his team around making the open ran network metaverse ready, which is essentially what I summarized. Low latency, high capacity, and also making the application network and network application. What we have done is for example, our layer one data comes in, we actually can make that available through API. That does a lot of inference learning that can happen, understanding what is happening in the most demanding fiscal layer. And then you can learn that, understand how the application perform and behave under certain [16:25 inaudible] condition, things like that is one of the things you could make metaverse ready network. 



But most important, we make sure that our roadmap, our Silicon roadmap can manage that type of a capacity, a low latency in a very power efficient way, including some inline accelerator we brought in, adding on [16:48 inaudible] and adding some of the additional instruction set just for this kind of 5g. So many things are happening because we do project that the network is going to be heavily handle, which you just described. I mean, that's just the path that network will take. 



Abe Nejad: So Mehran in your opinion, what needs to happen to accelerate, to catapult, the metaverse ready network?



Mehran: I think the things that would help accelerate it significantly is new and interesting use cases. Think about intelligent transportation or location of our services or medical use cases. Image you walk into the bus station, the bus already knows your destination, depending on the number of people standing, the bus gets read outed to maximize throughput of the bus, for example, or you go to the doctor, he's on your screen, he's touching, he's getting all your vitals automatically. You don't have to move anywhere. If you provide the low latency networks and the infrastructure to support it, all these use cases would happen. And when the use cases happen, that's where the revolution happens. 



That's when all these things become reality. And Facebook realizes that. You guys are all about connectivity. You want to connect people, and this is the way to connect people. It removes boundaries, removes nations, brings people together. The social value, the efficiency value of the infrastructure network would drive that. Its quite an exciting thing in my view.



Abe Nejad: So direct correlation between open ran and metaverse ready networks, is that fair to say?



Manish: Well, yes. I mean partly, because as I said, Metaverse has certain things that will require low latency, symmetrical bandwidth. I would argue, edge capabilities, open ran starts to become a key building block over time, because it does through that whole hardware, software disaggregation makes the underlying infrastructure more programmable. So there are a lot of very intelligent things that you could do in terms of managing the experiences. And also it makes the whole infrastructure more software-defined, which drives more innovation, which will be required to, again, enable metaverse at scale. So it's both those things. It's the disaggregation. It starts to put those basic pieces in place, but then to be clear, there's a lot of work. The industry as a whole has to do to enable meta work. And I just gave quality of experience as an example, but there's a whole more body of work, but yes, open ran puts those basic building blocks in place.

 

Caroline: Especially around, I just picked that one word that we'd be driving open ran is programability and softer, software defined. And I think this has been foundational metaverse network. What's metaverse network, ready network. You have to attract developers. You have to be able to program because of the continuous evolution of metaverse application. So that sort of things really requires programability, requires software define, and the ability to attract developers that can write applications on the open ran metaverse ready network as if they were writing something for Android or for iOS. I think those are really key elements that we as a open ran community need to focus on to make us metaverse ready.



Abe Nejad: It makes sense that those are the buzzwords or keywords that you're hearing on the floor here in Barcelona. Now that we've had sort of a little bit of a shedding of the spotlight on these two terms. So it makes a lot more sense to me at least now. So Mehran again, this is sort of our second session back to back that we've had you on. We appreciate your time and giving us your insight.



Mehran: Enjoyed it quite a bit. Thank you.



Abe Nejad: Congratulations on the news, by the way, yesterday with Rakuten Symphony.



Mehran: Quite exciting, being part of Symphony, I think there's a lot of synergy and we are kind of excited about the future.



Abe Nejad: Absolutely. And we'll have Rakuten symphony and Robin and some other folks right in this room tomorrow to talk about that. So looking forward to that. Caroline Chan always a fixture, if I may use that word, if that's okay on our programs. You've been extremely supportive of NMG over the years. So we appreciate. 



Caroline: Love seeing tou face to face.



Abe Nejad: And thanks for the gift, by the way, I got snacks or sweets rather. Manish it's been a while. You've changed course as far as shifting to over to meta again formally known as Facebook for those that don't know that. So it's got to be super exciting. I can only imagine. I guess the sky's the limit there.



Manish: It's definitely exciting. I think we've been working pretty much focused on open ran and enabling that and metaverse and getting from the connectivity side, metaverse ready networks. That really, really is an exciting challenge. So definitely excited about that and great to see everyone in person. This is my first event post COVID. So I'm really, really happy to be back and seeing you and everyone in person.



Abe Nejad: Well, I'm sure we'll see each other very, very soon. And thanks for your time again.



Thank you. Thank you. Thank you.

 

 


For any inquiries, please email anejad@thenetworkmediagroup.com

Abe Nejad5G RAN