Guy Daniels, TelecomTV (00:00:24):
Welcome back to the Open Ran Summit, part of our DSP Leaders Coverage and Time. Now for our live q and a show, I'm Guy Daniels and this is the second of two q and a shows. It's your final chance to ask questions on open ran from deployment strategies, Rick Evolution to upcoming developments. Now we've already received a lot of questions from you covering a wide range of open ran topics. However, if you haven't yet sent one in, then you do need to be quick. Please use the q and a form that's on the website. Well, let's now meet our guests who are going to help answer all of your questions. And joining us live on the program today are Rahul Atri, who is president of the OSS Business Unit at Rakuten Symphony. Beth Cohen, SDN Network product Strategy for Verizon Business Group. Ian Hood CTO, chief Strategist Global Industries at Red Hat, Richard MacKenzie, distinguished Engineer Wireless Networks with BT Matteo Fiorani, head of distributed Unit and infrastructure for Ericsson and Patrick Lopez who is the founder and CEO of core analysis.

(00:01:45):
Hello everyone, thanks so much for returning for our live show. As I said, we've got a lot of questions in from our viewers, so let's get straight to our first one. And here's the question, the adoption of Open ran by MNOs depends not only on technology or TCO, but also organizational realignment. As procurement and operational models are different to legacy networks, how is this work progressing? Well, Patrick, perhaps we could start by coming across to you for your views. It's not always technical with open ran, there's a lot of operational aspects. What's your take on how this is progressing?

Patrick Lopez, Core Analysis (00:02:27):
Thanks. Hey, I think Open Ran inherits a lot of other technology or if you will, there are a number of prerequisites in order to deploy open ran, you need to have a network that has a cloud native or a cloud abstraction capability. So you need a virtualization layer. You need to be able to purchase and deploy hardware separately from software and to have an open ran, sorry, a cloud native pipeline. So you need to adopt capabilities such as C-I-C-D-C-T. So it's a lot of new skills to onboard At the same time as you're deploying a new technology and possibly new vendors with open RAN and many operators, even if they were not sure how much open ran they were going to deploy over time, they took open run as a poster child or rather as a working environment in order to deploy and to learn all these new technologies, how to deploy commercial off the shelf hardware, how to deploy software in a programmable manner, how to use systems integrator to help them along those tasks. So I think that we're not there yet, but many network operators that have deployed Open, ran have learned a lot of those key skills that are accessory for the future of 5G and six G.

Guy Daniels, TelecomTV (00:03:53):
Great. Thanks very much Patrick. As you say, a lot of key skills that they need to learn and adopt. Let's go next to Beth and then we'll get some more views, but Beth next please.

Beth Cohen, Verizon (00:04:04):
Sure. I want to talk from the perspective of the telco and I would say that good news is that we are able to build on a lot of the skill sets that we've been using for our cloud adoptions and being able to use it for oan. So our operations people have gotten used to virtualized systems, Kubernetes there are actually much more familiar with these tools than let's say five years ago and so that they can actually build on that. And so I think that we're further along in terms of the skillset or rather the underlying skills that are needed to be successful with Iran. So of course obviously we need to build on top of that to add the specific skills for ran.

Guy Daniels, TelecomTV (00:04:58):
Great, thanks very much Beth. Good to hear that we are moving along here and as we've heard so far, I'm learning from cloud native as well. Rahul, let's come across to you for your thoughts as to how we're progressing.

Rahul Atri, Rakuten Symphony (00:05:13):
I think Patrick covered some of the points. Yes, the procurement cycle changes we need to adapt to have the decoupled way of buying in the equipment and technology. The second aspect I believe is what we are also seeing in the market. If you are moving to the software defined networks or cloud native networks, you have to be either a build versus buy organization. If you're a build organization like what we did in Rakuten, you build your own softwares, you try to build your own automation layers, you build your pipelines and you build the skillset to be able to manage the network which has multiple layers right now starting from infrastructure cloud and then application and services. The other kind of organizations are also buy which needs a lot of ence in the ecosystem. You need to find the right partners, you have to have the system integration skillset, not with the partner but also with yourself. I think if we can take the technological call on the right partner that is buy versus build the angle the otherwise is with cloud native networks you have to do things on your own. You have to become the technology company rather than just being the, so for me, I think the distinction and decision of becoming a build versus and finding the right model for build and finding the right ecosystem for buy is the change.

Guy Daniels, TelecomTV (00:06:37):
Oh, that's interesting. Thank you Rahul and Ian, let's come across to you now for your thoughts on this.

Ian Hood, Red Hat (00:06:42):
Yeah, so I'm going to follow up on what Beth made to comment that a lot of the customers have actually already learned a lot as they were building out their cores because on that to 5G core, they were building a cloud native deployment in that context and those skills and what I found is that the different operators had improvement in skill sets in different teams and then they kind of moved those skills around and that's kind of what's going on now is that the teams that really understand RAN are being cross-functionally joined if you will with the teams that own the platform and that's driving a faster approach and I think that that learning for some of our customers has actually made it easier for them to kind as Rahul was talking about, control their own destiny and be kind of the integrator leader, if you will with their partners. So this is kind of what I'm seeing.

Guy Daniels, TelecomTV (00:07:32):
Fantastic. Thanks very much for those insights. Ian and Matteo, let's come across to you as well please.

Matteo Fiorani, Ericsson (00:07:38):
Yeah, thanks Guy. My comments is somehow aligned to also what Rahul have. There are different models that the operators can choose and that would trigger different type of competence needed on the operator side as well. So we'd refer to something like an operator that chooses to get preintegrated solutions out of the vendors. Then the vendor can take part of the load and can act as the prime integrator on behalf of the operator. And then of course the operator needs to build some core competence still, but that level of core competencies doesn't have to be that high, so they don't need the very specialized organizations to be able to run an open run network. On the other hand, there are operators taking a different approach. They do more in-house and in that case they need to build much more core competence to be able to run them to add system. They need to have integration capability by themselves. So that would require of course a much bigger change on the operator side and much bigger organizations to be able to run the network.

Guy Daniels, TelecomTV (00:08:31):
Great, thank you Matteo and thanks everyone for those comments for that first viewer question. I do want to move on now because as I say, we've got a lot of questions and Richard, I think I'd like to come to you first on this next question if I may, is about power consumption and the viewer asks, power consumption still seems to be a weakness of open ran solutions which will not be addressed until a new generation of chip sets hits the market in about two years time. How do we address this issue in the meantime? Well first of all Richard, is there an issue?

Richard MacKenzie, BT (00:09:05):
Absolutely there's an issue with power consumption, but that's not specific to open run. A lot of the problems we're trying to address in the traditional architecture networks as well. And a lot of the key ways that we can save energy are the things that we can do that are not specific to open run. So particularly just turning carriers off at quiet times, that's all about turning off the power amplifier. That's the main consumer. So if we're looking at a power consumption problem, that's the first thing we should address and we can address that now and it seems like the question is more focused on the actual chip set the platform. So there are certainly energy saving features that address that already. There's a number of different platforms so we're trying to address in open run to make sure that we're not just running on one particular type of platform, but that we can do X 86, we can do arm, we can do other platforms.

(00:10:04):
We want to make sure that they've all got different trade-offs in terms of consumption. So there are features that can happen now I'm sure in two years there's another generation of chip sets that's got some other incredible breakthrough, but again, that won't be a one-off big step I'm sure every generation of chip set that comes through, one of the things we're going to be asking is how does it affect power consumption? So that's going to be a gradual thing and we're going to see those efficiencies as time progresses. So certainly it is one of the priorities right now the recent Rick Forum event, they put energy saving as one of the two prioritized use cases and I would say that that's probably the area where there's the most attention in open run right now.

Guy Daniels, TelecomTV (00:10:47):
Good to hear. Thanks very much for those thoughts Richard. And we're here from Rahul in a minute, but Ian let's come across to you first.

Ian Hood, Red Hat (00:10:55):
Yeah, certainly I agree that power consumption is a challenge across the entire architecture and many things can be done now as been said in terms of what we do with the radios and spectrum, but we've already been through a couple of iterations of the technology for the servers that are doing the DU and cuus and one of the things in the open community we've done is we've made it so you can actually use operators to tune the underlying hardware and then we've got the Kepler project that allows us to monitor the power consumption and then adjust it on the fly. So these are things that we can do and is now a softer defined approach. So if you do bring some new hardware to the party to support the next generation of massive MMO at some higher frequency like we're going to need at some point, then we'd again use a softer defined operator that actually is a standard way to actually reconfigure that hardware underneath the platform.

(00:11:46):
Then that's how we'll kind of deal with it. So I think we've kind of got it under control, but certainly more work to be done to automate how we change the power on the fly and decide where's the best power to be utilized across, not just the radios with Spectrum and what frequencies they're using and the interfaces of the CU and DU but across the entire infrastructure because one of the things that is still out there is the cell site router, right? That's still a physical device. And so once we look at how we solve for that problem solve for the rick as well to actually utilize that for power consumption, these things will kind of evolve as well to kind of keep driving down and making this whole thing efficient.

Guy Daniels, TelecomTV (00:12:26):
Great, thanks very much. And as we've heard over these two days, it's an evolution, it's an ongoing development process here. Rahul, I promise to come over to you. So what's your thinking on the power issue?

Rahul Atri, Rakuten Symphony (00:12:37):
I think like Ian covered, it's a good problem to have to be very honest. If it's evolution, you're not going to solve all the problems day zero. I think this is one of the most talk topic about whenever the TCO or the open ran topics comes out, I think we have done certain evolutions, we're able to understand the practice, we are able to, it grows with especially on the new application going forward, like was mentioning, there are some of open source projects coming out of CNCF. The project is the most interesting one where you can actually understand the P state C states see the CPU cycles, frequency utilizations of the chip set and also read I think around 120 different events and KPIs to understand how the application is performing. But on the other side it is also not only the hardware but also how your cloud is performing.

(00:13:26):
For example, we are able to tune the cloud onto one two core. We are able to insert a lot of agents in terms of PVPF or the Kepler project to understand how good is the application design, how you can do non pcards versus the P cars, and then on the management orchestration layer you always talk about Rick and the energy saving and other things. So for me I think it's a great problem to solve coming back to the earlier point as well, it's a problem of the ecosystem to solve together. It's not the application, it's going to be the evolution. Talking about similar ways we have seen in the past and it's why I say it's a good problem because we have multiple solution to that hardware chip set is one part of it. Application software, how it is written, how it manages and take the hardware into the loop is second part of it. And then the third option, Rick platforms all has to go together to solve this problem. But yeah, it's an evolution. We have seen the results we're going through, at least we know how and what is happening and then we have been able to say close the loop, understand take decisions and because it's software defined is much more quicker to take those decisions software turn on some of the carriers or some of the feature sets.

Guy Daniels, TelecomTV (00:14:39):
Great, thanks very much Rahul. Thanks for those comments everyone. I'll move on now because I've got a great question in here. Is there any space for a startup to play a competitive role in Open RAN and the Rick ecosystem and can we identify any likely areas? Beth, it's tricky for startups especially on the hardware and the infrastructure side. What's your thoughts on this?

Beth Cohen, Verizon (00:15:09):
I think there is room for startups. Verizon does support startups. We invest in them if they look promising and we understand that it's difficult to break into the ecosystem, but there's also real opportunity for startups to really disrupt the market. So the incumbents tend to be, they're looking at the bottom line and saying, Hey, I got to get my numbers every month or every quarter, but the startup is in a different situation, so they have, the thing about startups is that they have something that the incumbents don't have. They have a blank piece of paper, which means that they can really rethink how these systems work and how they can be delivered in a different way and that's very valuable. I know that certainly Verizon values that and we do invest in startups. I think there is, but it's done a little differently than the traditional venture capitals. I know that some of the incumbents will invest in startups as well. So yes, definitely they have a role in the ecosystem that can't be duplicated.

Guy Daniels, TelecomTV (00:16:32):
Absolutely. Thank you very much Beth. We've got a few comments to get through on this question, but Patrick, I'm going to come across to you next.

Patrick Lopez, Core Analysis (00:16:40):
I think historically startups have had difficulty penetrating telecom's market and it's not because there hasn't been enough innovation, it is just the nature of the business in the sense that network operators, in order to be able to deploy a technology at scale, need vendor that have the same scale. So what it means in many cases is that those startups, those that are successful, those that have brought forward a technology or a product that is disruptive and create value, they get acquired, they get acquired by a larger vendor and within the parameter of a larger structure they can scale better and then they can become vendor to the technology operators. Startups that have been able to basically grow and deploy at network operators and then grow organically are few and far in between. There are some that have been successful in that space, but in most cases a startup that has a successful product or technology, their path to growth is through m and a.

Guy Daniels, TelecomTV (00:18:04):
Yeah, absolutely. Thanks Patrick. We do hear this, don't we, about the roots for innovation that the startups can provide, right where we're going to go next. Rahul, let's come across to you next for your thoughts on the potential for startups.

Rahul Atri, Rakuten Symphony (00:18:20):
Absolutely, yes. I think one thing which the startups are very good at solve and solve problems differently, not just like what we do in telecom every day and out, we have worked very closely with startup in Rakuten, right from the platform building our own Kubernetes platform and the company happened to be outside telecom infrastructure, there were Kubernetes storage player and now they have the Kubernetes platform of our own. We also had a lot of innovation done on the databases side with Yoga Bytes, a lot of innovation done with Minai on the storage side. So I think absolutely yes, they come in with the tendency to solve problems, not with the telecoms or non telecom. They also bring the freshness. They also challenge the current ecosystem. The other part of the thing is whatever open source projects or software entities, we use all our brand built by startups in one way or the other.

(00:19:23):
I think C NCF F is doing great job bid with figuring out how to have a win-win enterprise model for these startups. We also have to do a lot more in telecom, nurture them, build them a part of the ecosystem to say that you're great with solving this problem and this is the larger picture, but you fit in here so come partner with us. I think a lot of operators are doing that, especially in terms of open Ran and Rick, there's an opportunity because it kind of moves the telecom into the space of technology becoming more software driven, become more like IT services. And then with Rick, some of the applications can be designed by startup. There are a lot of use cases which will come going forward to closed loop to do AI application to build small language model if not large to near real time Rick, and I'm not sure if we have some of the partners looking into already tapping in, but I'm very sure a lot of us enjoy working with them, learn a lot from them. It's just that how do we nurture them and make sure that it's a win-win for them. They will be bandwidth and the people to just have the long cycles, which we are not that famous for in telecom.

Guy Daniels, TelecomTV (00:20:35):
Yeah, quite right. Nicely put. Thank you very much Rahul, Ian, we'll get to you in a moment, but first of all I just want to go across to Richard.

Richard MacKenzie, BT (00:20:43):
Yes, there's a few things worth mentioning here. One is obviously across the globe there's a lot of funding initiatives to promote Open Ramp and a lot of those create collaborations where they are looking to make sure there's a role for some of the smaller companies and startups. So we are seeing them involved in a lot of collaborations and trials. I suppose the real challenge and the proof is going to be at the end of those activities, have they done something to actually win some business so that they're no longer reliant on that incentivized funding. But I think we're going to see a lot and we've seen particularly with the Rick, that's where the most obvious opportunity exists because that's the whole idea that with the Rick, if you can develop one particular X app or R app, you are no longer responsible for providing a full feature set that supports all services.

(00:21:38):
You've got your own innovation that focuses on one key benefit and you can focus on that and you just turn that into the product. So that really does enable an opportunity for startups and small businesses to be focused on what they're best at and not have to spread out their expertise over everything and become too weak. They can stick on being focused and being very strong. And I think a lot of the examples are university spin outs, but we are seeing other smaller companies, so particularly with the rick, we're looking at some vendors are now promoting themselves as xapp and R app vendors. So I think there there's going to be a lot of opportunities there and certainly it's not an easy win, but I think some of them are going to succeed.

Guy Daniels, TelecomTV (00:22:27):
Well, let's hope so. Absolutely. But you're right, it is tricky. Ian, let's come across to you for your thoughts on the potential of startups to play in open run.

Ian Hood, Red Hat (00:22:40):
Yeah, so as many of my colleagues have said, the startups have always been a part of the telecom industry and over time they as Patrick said, becomes an m and a piece and unfortunately for some that don't get funded the right, it's kind of a timing thing in anything as to whether or not they can actually survive to the point where they get acquired. But I think the areas where we are going to see them significantly more than we have in what I would call the, I'll call it the highway portion, which is the radio network out to the au. We've had limited new startups there, but certainly in the orchestration piece, the Rick piece and especially as Patrick was saying, the X apps to drive either location, other types of capabilities, and of course we have many startups and ISVs already in play in the AI space for all kinds of applications.

(00:23:33):
We're going to see them being applied into this equation to support the performance capacity sustainability equation that we talked about. So I do see lots of room for them, but again, it's a timing and as I say, if we don't kind of get this deployment moving fast enough, the timing may not be right for some of those startups who may have come along too early in the timeline and aren't funded long enough to actually get acquired. So it's a challenging kind of down the edge kind of equation for the startups, but I still think it's a good opportunity for them.

Guy Daniels, TelecomTV (00:24:06):
Great. Thank you very much Ian. Well thank you everyone for those comments. Before we get onto our next audience question, it is time for us to look at our audience poll for the Open Ran summit and the question we are asking you this week is, which are the most important areas of focus for the open ran community during the next 12 months? And you should, there you go. You can see the real time votes appearing right here. Interesting range of results so far. More field trials please. Total cost of ownership and operator commitment seem to be heading the pack but only slightly. Right. Well look, if you have yet to vote then you are running out of time. We will keep the polls open until the end of today and then we'll analyze the results on telecom tv. Right. As I said, we still have time for more questions, plenty of time and hopefully plenty more questions. So here's the next one. How will Open RAN impact the market potential for massive MIMO? Interesting question. We've had a couple of questions along a similar theme here. Mateo, I'm going to come across to you for your thoughts on this one.

Matteo Fiorani, Ericsson (00:25:26):
Yeah, massive mimo is and open run is an interesting question and I think they have massive mimo and Open run a long history. I think I will start with the standardization aspect. I think one important piece to be able to make open ran Mass mimus successful is to select the right architecture that can be competitive in terms of performance with existing systems. And I think a lot of concrete steps have been taken there by the OR alliance that have recently decided to standardize an architecture in is optimized for mass IO. This will make the radio a little bit more complex but would also give the possibility to build open massive MIMO radios that are from a performance perspective at least on par with existing systems. The specification now are almost ready. They will be published in the June release train from the Alliance. So I think that is an important step for open run Massive mimo to be successful.

(00:26:21):
Then another aspect that will be important is what kind of hardware availability will be there over time because that is quite important for Mam Mima a lot more than for classic remote radio units. You need massive MA radio runs a lot of software, so you need to have a powerful silicon for that and there is an ecosystem building up for that. There are many different silicon vendor building solutions for open run mass IMO and that is important to see how this will evolve and what scale they will get because scale of course, it's very important for these vendors to be able to actually compete in the market. So these two aspects, standardization and availability of silicone are key for open run mass MMO to be able to evolve and become fully competitive.

Guy Daniels, TelecomTV (00:27:05):
Great, thanks for those insights Matteo. Good to see the ecosystem developing for this. Patrick, what are you seeing in the market regards massive MIMO and how it will impact open ran?

Patrick Lopez, Core Analysis (00:27:18):
Thank you. I have maybe a little different perspective than my colleague from Ericsson here. I've seen massive IMO platform being deployed commercially at scale in networks and we've seen them in Japan for instance, either at entity Doomo or Rakuten Mobile. We're seeing them in the us, we're seeing them in Western Europe, we're seeing them in Canada. So I don't want people think that in order to deploy massive MMO in open ran you need to wait, you need to wait for more standards because the massive MO that are being deployed commercially right now of the same level and in some cases better level than some traditional vendors solution in the massive MMO space, it's not necessarily a problem of standardization. Standardization will continue to evolve and will continue to create more variants and more different ways to deploy massive MIMO. And I think that the uplink optimization that is being proposed and finalized at the open RAN alliance is a good addition to the tool set in order to deploy other capabilities. But I think that we see today massive MIMO being deployed commercially in open ran without performance issues. So that's the message I want to carry with you.

Guy Daniels, TelecomTV (00:28:58):
Great, thanks very much Patrick. A couple of strong messages there. That's good and thank you to our viewer for sending in that question. Well, here's our next one. Some telcos are open ran leaders, but others are happy to follow. How closely should these more cautious telcos follow and is there a danger that they could wait too long? Ian, let me start by asking you about this one and we know there's always front runners in any technology and open run is no exception, but what about those tokos who are just the more sitting back and seeing how things turn out?

Ian Hood, Red Hat (00:29:36):
I think the simplest thing to say is that if you will, it's time to just get started. We've had a couple of years now of work by the leaders around the world in Japan, in Europe and North America. So we're at to the point where we know that we can actually do this and we see the benefits now coming to those leaders that have gone there, but the longer you wait, the harder it is to actually kind of get there because there is, as we talked in this discussion, things to do with your organization, things to do with your operational teams. These are things that aren't really necessarily an open ran question. It's actually getting your teams capable of, and that's I think the biggest thing holding a lot of 'em back is their confidence in being able to be sort of the leader of their own destiny and take that risk to drive this forward to their actual goal, which the goal as we've kind of talked about is how do I actually increase my share of consumers in my market or go ahead and compete elsewhere and secondly to actually go after the enterprise market.

(00:30:41):
And so these things will continue to be delayed further if they keep waiting. So my sort of view is that the people who are first to market tend to lead and take the market share. That's always the way it's been in any industry and so there is a danger to continuing to wait and watch by the sidelines.

Guy Daniels, TelecomTV (00:30:58):
Great. Thanks Ian. So yeah, no time at the present. I think to delve into open run, Beth, you have a similar mindset to Ian on what those follower telcos should be doing now.

Beth Cohen, Verizon (00:31:15):
I think it really depends upon where the telco is. I personally think that the smaller telcos need to be fast followers and really step up the pace. Telcos at the end of the day are technology companies and I think we forget that. And so they love their bits, they love their bites, they love the widgets, and so I think that particularly the larger telcos that have larger budgets are more than happy to jump into O ran at the appropriate time and are willing and able to do that. The smaller ones tend to be more constrained by business factors, so for them they have to kind of dip their toes in the water, but I would not recommend that you wait too long. You really need to, I wouldn't say necessarily be a leader, but being a fast follower, that's really weird. That's the sweet spot for the smaller telcos.

Guy Daniels, TelecomTV (00:32:20):
Okay, great. Thanks Beth. Fast follower. Yeah, absolutely. We've got a number of other comments we're going to hear from, but first of all Richard, let me come across to you first.

Richard MacKenzie, BT (00:32:30):
Yeah, so I think one of the big opportunities missed with open run was it wasn't quite ready for 5G rollout. So for a lot of Brownfield operators, the real problem is that we launched 5G, we've already got existing suppliers for 5G. So the opportunity to then roll out any new technology is when that 5G rollout equipment comes to end of life or wherever we've got new areas that we want to deploy. So I think one of the early ways that we can roll out as brownfield operators is when our existing suppliers move their offerings to be more open run compliant. So that's one way to do it, but certainly when you say when we're going to just bring in a new supplier who does open run, you've got to wait until there's actually an opportunity to deploy something new. So I think that the factory missed 5G, that's put a lot of people back so there's no rush to deploy open run just for the sake of it. But there are a few other applications such as truck hosting, building and private networks where I think even Brownfield operators are looking to deploy what you would also almost call greenfield deployments.

Guy Daniels, TelecomTV (00:33:41):
Great, thanks so much Richard. As you say, there's a number of factors influencing deployment timeframes there. Matteo, let's come across to you next please.

Matteo Fiorani, Ericsson (00:33:51):
Yeah, I think that as always there is a trade off in what is best to do. I mean if you are a front runner and start adopting the technology very early, you wouldn't have to go through a maturation process because the technology needs to mature. A lot of that has happened in the past few year and some of the leaders in the industry in terms of operators have taken that cost. But it does come with the cost being a leader and adopting the technology first because all the issues for the technology need to be ironed out. So there could be a benefit especially for say smaller CSPs in not being on the leader front and wait a little bit longer. But there is a risk of course in waiting too long as well because as we discussed also earlier, there is a maturation process the organizations need to evolve. There are a lot of lessons that need to be learned in deploying a new system. So waiting too long means also being too late in the game and being too late in doing the transformation. So depending on the size of the CSP and the type of business they want to run, there could be a different time and different sweet spot on when to exactly start, but it's better not to wait long.

Guy Daniels, TelecomTV (00:34:57):
Great, thanks very much Matteo. Don't wait long then. Rahul, let's come across to you next please.

Rahul Atri, Rakuten Symphony (00:35:05):
It's so interesting. In the same forum we have discussed about innovation startups, technology, digitalization, ecosystem, and we actually started from offering opportunities to startups as well. So I believe that there's a shell around telecom. We are in a comfort zone and sometimes we wait for the next generation like 5G and 4G and five six G maybe to say, yeah, that's the right time to move along. I think we all know that we have to break the shell in telecom and move forward into doing something more, maybe open up new services on the service layer use cases, whether these are private network for manufacturing hub or a hospital or something. But the shell is that we need to make network more programmable in the sense they have to be platforms always available, self-configurable, runtime, dynamic, whatever we want to solve the use cases on the services layer.

(00:36:03):
If we don't do selling more data, selling more sim cards are getting saturated. That's why we see the arpo hit on 5G. I'm not sure it's the usual chance of having 3G, 4G, 5G or it's not working, even working, but I truly believe that we are at a juncture where we can, we are on the redefining telecom and we have the perfect opportunity to do that. I've said it before, anything which touches telecom, whether it is a delivery app, food app, cab, booking app, everyone has become unicorn on top of connectivity, but we suffer daily on terms of how to increase more arpu Open RAN is one of the ways to become more programmable in terms of network, whether it's starts with cloud. But yes, we will reach out to a point where we will have to do runtime slices, we have to assure quality of service, we have to do certain use cases which are totally not defined, not monetizable today, but the network has to become programmable and I believe that over has a great role to play along with Rick going forward along with other capabilities which will come going forward with open APIs and other things.

(00:37:17):
So I believe that if we all come together, create an ecosystem, don't have to solve the same problem. Again, different from BT is mentioning that in last two, three years the leaders have done a lot of fine tuning. We have had bad time, we have fine tuned, we have governed, we have figured out how to run these operations, why don't share with each other, why don't learn. And if you see that a lot of operator to build their own platforms. Now whether it is Rakuten, whether it is geo, we seeing entity also building something and there are a lot of forums which are coming together. So I think the ecosystem is right. It's all about some of the leaders and some of the operators coming together and assure and say yes, we will put the foot down and have open ran, deployed, whether it's private network, whether it is ruler, whether at the telecom hub. So I believe that the time is right and it's just about taking the right steps.

Guy Daniels, TelecomTV (00:38:15):
Great, thanks Richard. And other reasons there to jump into Open Ran and Beth, let's quickly come across to you for a pickup, a follow up comment.

Beth Cohen, Verizon (00:38:23):
Yeah, I just wanted to pick up on a comment about early adopters, right? So there is a risk when you are an early adopter. We're not a hardware company obviously, so we obviously purchase our hardware from our ecosystem and the network O ran and just ran in general is the ecosystem and the technology is changing very rapidly. So we do risk having to rip and replace the hardware that's out there and that's obviously it can be a costly risk so we hope that O ran will actually help that, but it's not a hundred percent. So I understand why some of the smaller companies might not want to take that risk of implementing something that ends up not working for them and needs to be replaced like a year or two down the road.

Guy Daniels, TelecomTV (00:39:31):
Yeah, absolutely. Thank you very much for those observations Beth, right. Let's move swiftly on to our next question, see what else has come in from the audience as CapEx reductions are implemented by the mobile operators, how will this impact the growth of Open Ran? Interesting question Patrick, are you able to shed some light and observations on this?

Patrick Lopez, Core Analysis (00:39:57):
I can try. Certainly total cost of ownership has been one of the early claims from Open ran in term of reduction and obviously the costs of deploying or owning an open run system seems to be fairly advantages compared to more traditional RAN systems. If you are starting from scratch and we've seen I think every greenfield operator in the world has selected open ran system rather than a traditional RAN system for the speed of deployment and the cost of ownership when you buy the system and deployed from scratch for traditional brownfield operator, it's a different story because, well just as it was mentioned earlier, you don't deploy a new technology or a new vendor just because they're available, even if they're better, even if they're more cost effective because the reality is that swapping a vendor or introducing a new vendor in the telecoms environment, particularly the random environment is costly.

(00:41:13):
Any which way you look at it, it's probably 40, 50 million over a period of two, three years to do that just in terms of training and processes of people internally. So you have to wait for the right window and as mentioned there was one window going from 4G to 5G, they're another window for some operators that have deployed maybe Chinese RAN vendors and that are now forced to swap them out. Another opportunity is for those operators that are maybe moving from NSA to sa, another opportunity is the introduction and rollout of private networks and all of those are kind of like natural times at which an operator is looking at their network and thinking, okay, if I'm going to deploy something new now is it the right time to do it? With the new technology and certainly in term of CapEx reduction for operator, which is cost control comes also the quest to increase ARPU and increasing revenue comes from new services.

(00:42:25):
So rolling out new services while decreasing investment means that you need to buy basically platforms that are more open, platforms that are less proprietary because they can be reused and open RAN fits into that model because at the end of the day if you buy commercial of their shelf hardware, we can use that for open RAN and you can use that for something else. At the same time, if you buy a cloud infrastructure platform that is not specifically for open ran, open ran will profit from it but you can use it for other things. So all these benefits if you want point to the fact that in an environment where you want to reduce costs open RAN can be a good option if you have one of those windows of opportunities that are coming for you in the next months or years.

Guy Daniels, TelecomTV (00:43:22):
Great, thanks very much Patrick for those observations. It's worth mentioning that TCO issues topped our poll last year and it looks like it may well do the same again this year. Ian, lemme come across to you as well for your thoughts on how the trend towards CapEx reduction could be impacting open ran

Ian Hood, Red Hat (00:43:41):
Well I think the general situation for the operators is that quite frankly they're trying to drive down cost across their entire infrastructure and they do spend a large portion of their money CapEx and OPEX on legacy systems. So the only way that they actually go invest in other things is to look at those investments and say, okay, where can I not spend money on some of the legacy things and kind of balance this thing out? So that's kind of how that kind of drives it, but Patrick made a good point about there's a lot of compelling events as to when do you get into this business and to time that time that window so you can actually go after those other services that drive the revenue side of the equation. And so it's a constant balancing act of how do you actually improve the operational efficiency, the business efficiency of your company in terms of how you actually do processes And in the cloud native 5G core SA and NSA, we've kind of gone through this, how do we automate that process and we're going to apply that to the RAN to kind of drive down what a cost so we can actually reduce this expensive cost we have in the traditional ran, which is many, many truck rolls and the fact that the current LTE network is actually very much fixed at that edge of the network, these are the kind of things that have to kind of look at that cost and see how we can actually adjust where do I spend my money and investment while I'm dropping costs and spend elsewhere in the system?

(00:45:15):
We've even got people looking at how they can take their cos that they've got and kind of refine how they're built because they get a lot of legacy stuff in there that just needs to start coming out of the ground. So these are kind of the things that need to be balanced in looking at that and it's just an ongoing evolution of the economics to drive better efficiency and drive down costs to get to those new innovative services for the customer, both consumer and enterprise.

Guy Daniels, TelecomTV (00:45:43):
Great, thanks very much for those insights. Ian and Rahul, have you got a comment on this one as well?

Rahul Atri, Rakuten Symphony (00:45:50):
I think I support what Ian and one is the timing aspect of it. The other part is how do you start doing bit by bit? For example, in our case when we were doing and we were lucky that we had the Greenfield network obviously, but the challenges were just the same. You don't get the skill, you don't have the platforms. When we started in 13, there was not even the NFEO standards properly defined and call first not defined how to bring on a split seven based open RAN automation coming from multiple vendors and how to even deploy cloud. So I think it also circulates what you see the future architecture and evolution towards your teams organization and platforms. Even if you start slicing today, there are a lot of legacy platform which are available which even the organization don't think about open ran, they are trying to migrate away from it.

(00:46:47):
For example, unifying the observatory systems or unifying the OS system. People EMS going away or unifying them into one. Then automation platforms teams are even without the open ran networks, building CICD pipelines, building the automation workflows. Workflow is a workflow whether you trigger for the open ran systems or you trigger it for the usual closed loop automation system teams have actually worked upon. And RF engineering is all about writing those scripts, macros the optimizations and understanding when this KPI hits what next three KPIs to monitor and then make the changes on tilt, make the changes on, make changes on profilings and also monitor pre post. I think if you look into the operator ecosystem right through plan, build, operate, monitors secure, there are possibilities which already are hinting towards migrating towards the future architecture and future platforms. If you think about tier ones, most of them including Ericsson, Nokia have evolved and shown architecture of the feature where a common Rick platform can handle non-cloud native or Brownfield network, which is non-open run and also an open run.

(00:48:03):
All these SA applications are hinting towards moving to our non realtime R platforms. All the automation platforms talk about orchestration, whether you are talking about the legacy RAN or we are talking about the open ran deployments and as I said, C CDs, all pieces around the open RAN or the non open RAN radio networks are coming together and as organization you have to take those calls to invest in systems because even if you take a call on Open ran to be later on at least the platform, the ecosystem is ready, the skillset is incubated. People start thinking about technology more as software rather than networks and only the telecom part of it. And also you bring in that culture of building things, going integrations, understanding the details of it, understanding what the API payloads are and how they're generated, how the business rules are rated. And whenever the organization defined to bring the possibilities towards moving to Open ran, it is easier integrations and by that time you're already become a software native. Telecom.

Guy Daniels, TelecomTV (00:49:13):
Great ri, thanks very much indeed for those comments. We've got to move on the Rahul, we've at least one more question in if not two because we've got so many in the audience. So I'm going to move across to our next question which is do you believe Open Ran and Rick is a good fit for 5G Ran sharing amongst CSPs? Interesting take there Richard, can I come across to you for your thoughts on this one?

Richard MacKenzie, BT (00:49:40):
Absolutely. Ran sharing is one of the use cases that we think is going to be successful early on in the adoption of Open Ran site sharing already exists and it's there to help with cost savings, but it comes with some major drawbacks and one of the key drawbacks is you are essentially making decisions on investments with your competitors and in particular once you've established a site sharing situation, when you want to upgrade that site, you're waiting for your competitor's permission to upgrade. So it is never been ideal and this is where Open Ran really stands out because now we can share a platform, a hardware platform, but we can still make independent decisions on the software and if it does have some sort of impact on one of our sharing partners, we can quickly roll that back anywhere. So it does allow for a situation where we can share infrastructure very effectively, but the operators can still compete against each other.

(00:50:40):
So this is a really good opportunity and the UK has been very keen on this. So the four operators in the UK have formed the Jots consortium, so it's joint operator technical specifications and they've issued a number of technical specs that describe how a neutral host has to build a system that the operators are then prepared to connect back to their various core networks. So I think in the UK in particular, this is a good opportunity but I think this is also becoming Global Interest Small Cell Forum have been trying to promote what Jocks has been doing for a number of years and now we see recently Telecom Infra Project has launched a neutral host group. So I think we're going to see a lot of this in the near future.

Guy Daniels, TelecomTV (00:51:26):
Fantastic. Well we'll keep our eyes open for that. Thank you Richard. I do want to squeeze in one final question that we've just had in the past five minutes and it's a bit of a left field question, so if anyone can answer this one, I'd be delighted. If you can keep it short, I'd be even more delighted. And the question is, can open RAN technology be used in cellular environments? Now we had about a month ago, we has a comment from a wholesale operator hoping that we'll get some trickle down technology benefits from technology such as Open ran especially into the wholesale business. But I'm curious, I like the question, can we use our learnings from Open RAN or some specific technology in Open ran? Can we use that in cellular environments? Anybody optimistic for that? Ian, thank you. Let's go to Ian.

Ian Hood, Red Hat (00:52:20):
I'm going to follow up on what Beth made to comment that a lot of the customers have actually already learned a lot as they were building out their cores because on that to 5G core, they were building a cloud native deployment in that context and those skills. And what I found is that the different operators had improvement in skill sets in different teams and then they move those skills around. And that's what's going on now is that the teams that really understand RAN are being cross-functionally joined, if you will, with the teams that own the platform. And that's driving a faster approach. And I think that that learning for some of our customers has actually made it easier for them to kind of, as Rahul was talking about, control their own destiny and be kind of the integrator leader, if you will, with their partners. So this is kind of what I'm seeing is incubating your teams to actually go apply these processes and technologies, the CICD, those kinds of automation techniques to your business. Those are the other areas that this evolution of open ran can drive into other areas that are non, if you will.

Guy Daniels, TelecomTV (00:53:27):
Fantastic, thank you very much Ian. And we've got a couple more brief comments in if we can Beth next please.

Beth Cohen, Verizon (00:53:33):
Yeah, I think that the whole disaggregation lends itself to being applied to other network technologies, not just ran technology. So yeah, I definitely see that it's a revolution or evolution either way in the industry to really drive that. Once the software, it depends upon what layer you're at, but if you're at layer three, it shouldn't make any difference what's running underneath. And even layer two one could argue that there's a lot of opportunity for desegregation to really allow us to use it for optical networks and copper networks, although they're kind of going away, but certainly optical networks as they, at the end of the day, we still have an optical network in the core and that's not going to change anytime soon. But if we make sure that these networks works seamlessly with each other, that just solves a whole lot of problems. Right now the telecoms are saddled with the fact that our older wireline networks run fundamentally differently. That should not be the case. So I think there's a real opportunity to really make them more transparent across the networks.

Guy Daniels, TelecomTV (00:55:00):
Great. That's very encouraging. Thank you Beth. And some more quick comments in please Patrick, let's come across to you next.

Patrick Lopez, Core Analysis (00:55:08):
Certainly thank you. I was very fortunate to be at the RIG forum earlier this year and I think that in the R we can see illustrations and examples of how the technology can be applied in other fields or at least in other non cellular bandwidth. Management particularly comes to mind, find a proof of concept of modulation of frequency of the radio frequency from the telecoms network and from a radar application from the Department of Defense where basically you could detect the radar frequency and the spikes and have a collaborative method between the telecom network and the radar to use basically the bandwidth and the frequency in a way that is not adverse, where there's no need to mitigate interferences. And certainly if you think about that, you can think about wifi and you can think about other non cellular technologies and I think there's a lot of progress that can be made there to make sure that the technologies that we see today in the ran, even traditional ran like BIM forming, can be applied to other propagation techniques.

Guy Daniels, TelecomTV (00:56:38):
Thank you very much, Patrick. Yes, the Rick. And before we close this show, let's get a couple more quick comments in Rahul, let's come to you next.

Rahul Atri, Rakuten Symphony (00:56:49):
I'm very optimistic what Patrick said about Rick internally in Rakuten we actually call that as Nick sometimes, which is network intelligence controller white has to stop at radio itself. We can think about optimizing transport, optimizing or optimizing a lot of other applications as well. So I think the concept of having a common platforms and able to build runtime applications, small containers to do quick jobs can be taken into any other industry per se. Not even cellular or wireline networks, but the concept you can have a common platform to build automations and close to functions which are self-monitored, which can even roll back when it requires, which can even have a lifecycle of their own. The applications are very similar to what cloud native technology is called, function as a service similar to what AWS does with other places. So Rick is something which has the capability to enhance the common capabilities of building platforms on automation side workflows, all those things can be taken into the other industries as well. I think very optimistic that it's high time that also telecom and enterprise and it can come and more innovation flows from telecom to the other side of the world.

Guy Daniels, TelecomTV (00:58:08):
Yeah, let's hope so. Thanks very much Rahul. Great comments there. This is going to be our last question, so let's hear now from Richard.

Richard MacKenzie, BT (00:58:17):
Yeah, I think that the key term is programmable networks. So the fact that open ran gives us that programmable radio access network means that convergence, which we've been talking about for years between fixed and mobile, that's more realizable. And so particularly the fixed network and the wifi networks can all be converged because we've got this programmable network. And so yeah, I like that term network intelligence controller. That's very useful. How I start using that myself, I think that's certainly one of the key takeaways. And similar is what Patrick was saying at the Rick Forum. One of the panel sessions that was actually set up there, they were talking about dual use because it was all about the components that are being developed now with the mobile operators requirements in mind. A lot of that can be repurposed in that case for military applications. So that the US government were very interested in that and that's a completely different network. I think certainly for our interests at bt, we're looking at private networks and a lot of those private examples, you wouldn't really class as cellular, it's just wireless. But we're taking advantage of those common components developed for open RAN that can be used to build these dedicated networks.

Guy Daniels, TelecomTV (00:59:33):
Great. Thanks very much Richard. It is lovely to see us all sharing ideas like this. It's fantastic. Well, we are out of time. That was a long show. We are out of time now though. So thank you to all of our guest panelists who joined us for this live program and that is a wrap for this year's open RAN summit. Thank you to all those views who submitted questions. We tried to cover the most representative ones we could in the limited time we had available. Now, if you missed any of our programs featuring all of these industry experts, so many of them, then you can watch them on demand from our website. Our summit series takes a break for a while. We return in September with the cloud native Telco. However, next month sees the return of the DSP Leaders World Forum and we will be live streaming the entirety of the two day conference. So please make sure you register for that. Until then, thank you for watching and goodbye.