Ray Le Maistre, TelecomTV (00:25):
Hello, you are watching the Green Network Summit, part of our year round DSP Leaders Coverage. I'm Ray Le Maistre, editorial director at TelecomTV. Today's discussion looks at improving network energy efficiency with ai. It's one of the hottest topics in the telecom sector right now, and there are lots of different views on how and when this might be achieved. Recent studies suggest that energy costs account for on average, about 23% of a telco's network operating expenditure. But can telcos use AI and analytics to reduce costs in an efficient way without significant investments in additional network technology and applications and without impacting their services? Well, during today's discussion, we're aiming to identify some of the key approaches to the use of AI tools in improving network energy efficiency and get a sense for what strategies are worth pursuing. And I'm delighted to say that joining me on the program today are Diego Lopez, senior technology expert at Telefonica, and an Etsi fellow Mirko Voltolini, VP of Innovation at Colt Technology. Neil McRae, chief Network strategist at Juniper Networks, and Beth Cohen, telco Industry Analyst at Luth Computer Specialists. Hello everybody. Good to see you all. There's a lot of ground to cover and I'm certain that our audience will send in additional questions for our live show. So let's get started. Looking at the planning and build stage of a network, how can the use of AI specifically to improve energy efficiency be incorporated into this particular process, either for new sites or upgrading older ones? Neil, let's come to you first.

Neil McRae, Juniper Networks (02:19):
Yeah, hi Ray. So I mean, look, this is a bigger challenge. You start with a framework. What's your framework to look at plan and build? And then it's largely the simplest of things, which is how do you design your core network sites? How do you design your tower sites? How are you planning the infrastructure throughout your whole network? What advances in technology are you using in terms of cooling? How are you using AI to actually help you plan the site? So there's very many more AI based planning tools that will help with airflow. It'll help with designing the power, which is people, when you think about telco equipment, they think routers and switches and things like that, but actually it's all the power equipment. It's all the air conditioning, and there's a lot of AI tools out there. Actually, a lot of them come from the data center world where clearly they've got probably a big, slightly bigger challenge than we have in the telco world, but those tools are being imported into telecommunication companies because actually some of the challenges are similar in terms of network nodes and network sites.

(03:34):
And then I think the other thing, which probably come on to later on, which is ensuring that you've got good visibility of the infrastructure that you've got. One of the biggest challenges I hear from network operators, especially in how do we optimize the legacy, is how do I know that I'm counting everything? How do I know that I see everything? And that remains quite a big challenge, not just in this space but in many other spaces, but really engaging, pulling the data from the network, pulling the data from cooling, pulling the data from energy usage, trying to understand where you've got gaps, and then working with these tools to efficiently plan things. And then also, I mean, we even see some operators thinking about where they locate their sites in terms of location of even the buildings so that they maximize on things like solar power or they maximize on other environmental things that help with running the site more efficiently.

Ray Le Maistre, TelecomTV (04:36):
Okay, thanks Neil. Great point there about inventory and these AI tools coming from the data center sector. Mirko will come to you next. And then Beth, so Meco, how do you see the role of AI tools playing in this planning and building process?

Mirko Voltolini, Colt Technology (04:54):
I think if you look at the level of complexity we have in networks, and I'm thinking about our network, we operate a global network with hundreds of sites and thousands of links, several hundreds of sites. It has become impossible to do optimal planning. And I think the simplest use case, the one we are adopting today is for AI is demand forecasting where you can actually use AI to predict where traffic is going to go. And with that, then you can overlay information regarding where energy is more efficient is also greener, and combine this information to ensure optimal deployment of infrastructure. So I think that's applicable to both existing sites as well as new sites. But you can actually do this in a development forecasting type of approach. I think what we see this evolving into more is more of a digital twin type of model where you can create a duplicate of your network and using these techniques as well as ai, you can apply these forecasting techniques to what if scenarios. And with that then you can simulate what would happen if you apply certain scenario, what kind of energy consumption would you have, what kind of level of energy efficiency you're going to have, again with multiple scenarios. So it's basically using AI as a predictive tool to provide insights into potential inefficiencies and optimize the development.

Ray Le Maistre, TelecomTV (06:29):
Yeah, absolutely. We're hearing more and more about digital twins these days. And Beth, let's come to you. What role are you seeing for AI in this network planning and build stage?

Beth Cohen, Luth Computer Specialists (06:42):
Well, I'd like to pick up on something that Neil said about using the cloud vendors as kind of models. I think the cloud vendors have it simpler actually because they tend to be concentrated in only a few sites. They can in fact pick and choose to be next to a power generating dam, for example. So they can take advantage of low cost power and be more energy efficient in that way because one thing is energy efficiency equals cost savings. So I think that's part of the reason that telcos go after it. And of course any industry, but we have it, the telco industry actually has it a little more difficult because first of all, many of the sites have been around for decades and they've been optimized for other reasons. We also are running a wide area network. So sometimes the locations are set by things outside of energy considerations. So that makes it just that much more complex to drive that energy efficiency. And I think that's particularly important at the edge where we have very, so then at the edge we need to think about using more efficient hardware, more efficient operations, and that's where we can use AI to optimize at the edge. And obviously you can use it at the core as well, but I think it's going to provide more dividends at the edge.

Ray Le Maistre, TelecomTV (08:33):
Okay. Thanks Beth and Diego, let's come to you next. What about this early stage in the network rollout and even upgrade to sites to bring down this energy efficiency overhead?

Diego Lopez, Telefonica (08:48):
I think that this is something that we have to take into account as well is related with the fact that you have to make room for the AI itself in the new deployments or when adapting existing ones because you had to run, if not necessarily the ai, you had to run the data collection and some kind of data pre-processing to make it useful for the ai. And in many cases, the AI is in themselves because the degree of automation that you can achieve is very, when it comes to energy consumption is something that should be quite local in the calculations and with a short loop for decision. So I believe that this is an aspect that we have to take into account and rely on the fact that we have the advantage with the evolution of the network infrastructure to more and more cloud native, virtualized, et cetera. There is the possibility of having a balance of where you run the functions and where you run the ai and you can plan for this to be well balanced and well structured without making a significant or dedicated investment on dedicated hardware for simple models that would help you to save energy at the edge or a local point of process. Probably you don't need big GPUs or whatever, but simply a well-trained model and trustworthy data.

Ray Le Maistre, TelecomTV (10:27):
Yeah, no, absolutely. You've got to consider all the angles here in whether you are actually making any gains by using AI in this process. And Neil, did you want to come back in here with an additional point?

Neil McRae, Juniper Networks (10:40):
Yeah, I, and I agree with everything shared. I think the point I would also wanted to think about was just kind of layers in the network. We as telecommunication operators have always had these layers. And in modern networks with we've got 800 gig coherent optics that can enable you to take layers out. And when you take a layer out, that has a massive impact on your energy consumption. So instead of having maybe three or four times the number of optics, you can cut that down by maybe 50%, maybe even more. And we have customers at Juniper that are doing that with our new PTX platform and taking those layers out, it doesn't also, it saves you not just in energy, but it allows you to get upgrades done quicker, faster and bring new features to the network. So we see many network operators, and we coined this term from one of our customers in the Middle East doing this network modernization that's really using the best of breed capabilities that you can get today and rolling them out in a way that makes a substantial difference in the whole energy usage of the network.

(11:51):
And then the second point is we talk about data. Actually in this scenario, most of the data is in our engineer's heads that are on these sites. And I really encourage everyone to go in as you're building that data set for your ai, ensure that you're interviewing and including the people who work in these sites day by day because they've probably got the most juiciest knowledge about the site, what uses power, what are the challenges that they face? And that data is often overlooked because we're trying to pull it from systems. I really encourage people to go out and talk to on the ground experts in each of these locations because they've got gold in their heads that if you extract it, you can make a massive difference in the modeling that you do.

Ray Le Maistre, TelecomTV (12:35):
Okay, yeah, fantastic point. Mine the gold in those heads, that's what you want to do. Absolutely. And Beth, we'll come back to you for the final point on this question.

Beth Cohen, Luth Computer Specialists (12:46):
Yeah, I just wanted to pick up on what Neil said about the gold in the engineer's heads because it gets back to observability, which is if you're not looking for that in the systems, you're not going to find it and you're not going to get that information and that information is critical. And of course AI doesn't work if it doesn't have the information. So yeah, the engineers really do know what's going on and yeah, you definitely need to pay attention to what they know.

Ray Le Maistre, TelecomTV (13:20):
Okay, thanks Beth. So we've just looked there at the build and planning stage, but in terms of network operations, how is AI being used today by telcos to help improve energy efficiency and to reduce network power usage? Diego, let's start with you.

Diego Lopez, Telefonica (13:40):
I think that well at least let's say tier one operators are applying it not massively and probably not in substituting the human loop. The AI is being used for sure for planning. Mirko was mentioning this idea of demand forecasting. We are using this currently and it's something that is being used and in some cases we are achieving. Some time ago I had the opportunity to work with some colleagues that were making work on AI to optimize, not to optimize the use of the network, but to optimize the amount of energy that we were spending with the track roles when attending, dealing with incidents at customers premises. And it was an amazing amount of energy that we were saving informs probably of fuel for the trucks and the vans, but we were saving energy as well. So the idea is that I think that right now we have AI systems that are being applied here and there for different activities that are mostly focused on assessing decisions that are taken by engineers and planners.

(15:09):
Still, we still like the point in which we take the following step on making AI more part of the control at the top of certain control loops and trying to understand this. And something that we have started as well, and I believe is quite interesting, is trying to mix several different goals in the AI or using the coordination of different AI models to consider, for example, that well you can achieve a particular very high energy saving with certain decisions on routing or on the devices of the configuration of the devices that are active. But at the same time you have to evaluate which is the impact that has not only in particular on the natural stability on the network characteristics as a whole, but on the user experience in which AI is again an invaluable source of predictions. And there are some colleagues of mine that are working on this and they are achieving a quite interesting results on the balance between savings and while keeping a good user experience.

Ray Le Maistre, TelecomTV (16:31):
Yeah, absolutely a vital consideration there. Mirko will come to you next about the current use of AI in energy optimization and then we'll come to Beth. So Mirko, let's hear from you.

Mirko Voltolini, Colt Technology (16:46):
So I like to mention a specific use case, which is the one of optimized routing I mentioned earlier we have a very complex network and actually spans multiple countries and you I'm sure everybody's aware of the price of energy is being heavily dependent on the type of power plants you have. And it can vary quite a lot country by country there are actually some very large differences. Sometimes it's twice as much as expensive kilowatt per hour in a certain country compared to another one. And again, with our complexity, we have multiple options from going from A to B for setting up network routes for ourselves, for internal use, I mean operations and also exposing that to customers. So we have the ability to create a green paths that doesn't necessarily mean the shortest possible path, but is actually the most energy efficient path. It may not be again, necessarily the lowest possible of energy, but more the actual, again, lowest possible spend based on the path that you choose. So that's something we can use for our own internal optimization and also expose to customers we actually built into our NASA platform, the ability for customers to see and choose which path they want to go through based on the metrics they see about energy consumption.

Ray Le Maistre, TelecomTV (18:16):
Okay, great. Thanks Mirko. Great to hear real use cases coming through now. And Beth, you've been looking a lot into the use of AI tools to optimize energy efficiency at the edge of the network. I understand.

Beth Cohen, Luth Computer Specialists (18:33):
Yes. So I'm currently working with a group with the open infra foundation. We're the edge working group, we're working on a white paper on this topic should be out in a couple months. So what is immediately apparent with our initial conversations is just how complex the issue is. And I think it's a good use for AI because there's multiple variations and variables and factors as Mirko and Diego both mentioned that go into optimizing the network for energy efficiency, but also balancing that with making sure the network is available at the same time. I mean a lot of data centers, cloud applications optimize their energy efficiency by tracking the usage and then just turning machines off and then spinning them up when they're needed. But telco usage doesn't really follow that same very predictable path. So it requires a lot more understanding of the variables and that's where AI can help with the decision-making process. And of course, as Mirko mentioned, price of energy change is quite variable across different countries. Political factors come into play as well where one country may be pushing one type of energy over another, so the complexities are just multiply and that's where I think AI can really help.

Ray Le Maistre, TelecomTV (20:27):
Okay. Yeah, absolutely. Nothing is ever easy it seems in this industry at all. And Neil, let's come to you finally on this about the role of AI in helping with any energy efficiency in current network operations.

Neil McRae, Juniper Networks (20:43):
Yeah, Ray, so look, operations team, their goal is to minimize change because change is typically what causes outages. So I see lots of our customers using AI on things like configuration management. So we see many times customers saying, Hey, our power utilization's high, and we go and look and do some audit work and we find that often customers haven't turned those power saving features on. And that in itself is something very simple to fix, very simple to do, and we're encouraging our customers to use AI as a way of managing configurations and changes, but also looking at how we optimize those configurations to get the most out of the network. And then second and relate to that is how are you ensuring that the data is in the AI model and how are you ensuring that your inventory is a hundred percent up to date?

(21:49):
I mentioned this earlier, but increasingly we find the AI is comparing the bill and the usage compared to what it knows about the site. And it will flag, hey, there's something missing here because in theory this is what the bill should be, but the bill from the power company is this or the usage is this. And in ensuring that you go back and pick up on that because that in itself usually points to something that might be legacy that actually you don't even need and you can turn it off. And the simplest way of saving money in many cases is what are the things that are there that we just aren't using anymore, but we just haven't gone round and turned off. And again, an operator here in Europe had a kind of hackathon that was discover things on the network that weren't doing anything or weren't generating any value and then just go turn them off. Now there's some risk to that, so you have to follow process, et cetera, but one round of that can save you more than any other program you've got because it takes time to deploy these things. So really encouraging that AI based configuration management to ensure that what you've got out there is what you should have there and that isn't there, or anything that you don't know about that you're flagging that and you've got someone looking into it.

Ray Le Maistre, TelecomTV (23:11):
Okay. Yeah, great points, Neil. I mean this accuracy of data as a starting point is absolutely critical, and as you point out, these tools can sometimes highlight where there is a lack of that accuracy. Well, let's move on now because another thing that some industry organizations and operators and vendors have been looking at for a while is how to measure energy efficiency. So is there a role for AI in helping with metrics and measurements of energy to help telcos better understand the data? Neil, let's start with you.

Neil McRae, Juniper Networks (23:51):
I think it's a tough question, Ray. I don't see anything coming out of the world that's saying this measure is a sign of you're doing things right now, obviously we have all the measures that are being driven by government in terms of type one, type two, et cetera, and they're helpful in terms of thinking about the overall program that you've got, but I think we're still at the early days, particularly when it comes to AI in terms of what those standards are. One of the things that we are working on at Juniper is actually in the routing standards. So today's routing is very much focused on shortest path and you build on what Mirko said is sometimes shortest path isn't the best path from a energy utilization or even cost perspective. Usually it is with cost, but sometimes it isn't. And what we've been working on is how do we expand those standards so that we're measuring, we're taking a new parameter, which is the energy utilization or the green factor of any network choice, and then using AI to really go out and measure that, say, actually is this the best path for the network and one of our platforms?

(25:11):
Paragon enables you to do that kind of visualization. And as we define and work with industry on those standards, I think we'll see many more measures we'll start to understand and align on. But I think right now we're probably at the start of that journey, but lots of people across industry, including ourselves are working on that.

Ray Le Maistre, TelecomTV (25:36):
Okay, interesting. Beth, are you seeing any particular developments here using AI to measure energy usage in networks?

Beth Cohen, Luth Computer Specialists (25:47):
I'd say it's the early beginnings of it, and I think partially it has to do with that the biggest energy savings is in the hardware, and that has not been a priority of the hardware manufacturers. Obviously Neil Juniper is clearly getting the message, but traditionally it has not been a high priority for the vendors, and I think now that there is more of a priority, I think we'll start seeing better tools that will, again gets back to observability. If you're not looking for it, you won't see it. And so I think that there's lots of opportunities for AI to be applied, but we need to pull the data to get the data. AI relies on huge amounts of data and if we don't have the huge amounts of data yet, which I don't think we do, I think AI has limited use, but I think that's going to change in the next 12 to 18 months as we put a higher priority on improving the efficiency of the hardware itself, improving the software that runs on it and adding observability, adding ways of observing what's actually going on in the network.

Ray Le Maistre, TelecomTV (27:20):
Okay, great. Thanks Beth and Diego, let's come to you.

Diego Lopez, Telefonica (27:23):
I believe that using ai, one of the real challenges here is not only about running the network, but is about being able to evaluate which the price that you're paying in terms of energy consumption when providing a service. And in that case what you have is a big huge problem of data aggregation and of identify or separating data that is in many cases, provided as a whole, if you have a big optical switch that is working on a hundred gigs or several hundred gigs, et cetera, you have a lot of services that are going through and you would like to understand which is the composition of that energy consumption. How can certain actions may impact one service or the other on the one hand, on the other hand is to make your customers aware because this is an information that they keep asking and as the concern about the energy prices on the one side and all the concerns about the excess of energy consumption or becoming greener in general as society grows, the idea of providing accurate information so people know what they're consuming, how they're consuming it is something that is important in these identification of the data sources in this separation of data aggregates and aggregation in other packaging that would be useful for users to make informed decisions.

(29:08):
I believe there is a great field for the application of ai, and this is something that Neil mentioned standards before. For example, we have recently started an activity in the ITF that has a very original name. It's called the Green Working Group in which that we're trying to identify which are the data, how data can be aggregated, how can be assigned in temporary units, et cetera. This is something that we are dealing with to better understanding what we need and where we can apply ai.

Ray Le Maistre, TelecomTV (29:51):
Okay, great. Thanks Diego. The green working group, it does what it says on the tin, so that's as good a name as any I think right there. Okay, thanks everybody for that. Now let's have a broader look at the ecosystem here because there's a lot of expertise out there beyond the realms of the telecom sector. So how can telco's look beyond the usual ecosystems to create new partnerships that use expertise from academia, data sciences, energy companies, AI specialists and so on? Is there a new AI based ecosystem emerging that will help with the green network? Meco, let's come to you first on this.

Mirko Voltolini, Colt Technology (30:38):
I think it has been said already, but one of the biggest challenges we have here is data collection, getting good quality data. So I think the area where we can leverage expertise that is typically not something that telco sector has been leveraging a lot in the past is this area. If you look at how we've been operating network, typically data has been collected and then thrown away. So we haven't kept good historical data across the board and I think there is a treasure of data that we can collect. So leveraging tools and systems and expertise in this space I think is the biggest opportunity for us.

Ray Le Maistre, TelecomTV (31:21):
Okay, thanks Mirko. And Beth, how do you see the ecosystem widening, if at all?

Beth Cohen, Luth Computer Specialists (31:28):
Oh, I think it has to widen because telcos are really good at running networks, but telcos traditionally have mostly been consumers of energy. And yes, there's been some energy efficiency initiatives, but a lot of the are relationships with the energy companies have for the most part, hey, how much are we going to pay for electricity? And I think that we need to take a broader view of, because the energy companies know how to produce energy, they're looking for new ways to produce energy and be more efficient about it and lower costs, and we need to be there with them and be in partnership with them because we are obviously one of the major consumers of their energy that they generate. I think another area that we should be looking at standards bodies obviously is important and standards bodies, not just the telco traditional standards bodies that we normally work with, but the standards bodies that are part of the data science standards bodies, the energy utilities, standards, bodies as well as academia.

(32:54):
I think there's a lot of opportunity that we miss out on academia. I don't think the telco industry traditionally does much with academia, but as part of this white paper I'm working on, I discovered there's a whole lot of people working on energy efficiency in academia writing their thesis and doctoral dissertations on this topic and doing great work in the labs to really optimize energy efficiency and let that drive the decision making process, which is again, not traditionally how telcos approach things. So I think that we should really take advantage of that broader ecosystem and so that we can optimize the delivery of our services both cost effectively, energy efficiently, as well as optimal performance.

Ray Le Maistre, TelecomTV (34:02):
Okay, thanks Beth. Neil, how do you see the ecosystem spreading in this regard?

Neil McRae, Juniper Networks (34:09):
I don't see it spreading hugely, Ray. I mean I think everyone in the digital world is bringing on AI experts and I think it's about pairing your AI experts with your network engineers and doing that kind of self-learning shared learning process so that the tools and the systems that you're building are actually driving the right value. One thing I would say about the ecosystem that's there today but probably doesn't have the right priority in terms of management focused time, which is every telco spends millions, probably billions across the whole industry on power and cooling infrastructure. How many telcos look at that as a strategic investment? And I can tell you the answer, not many, but actually the way that you are managing that power build, the way that you are managing that cooling build can give you a massive uplift in energy savings. So rectification going from legacy rectification to S wave based rectification, that'll knock maybe four or five percentage points in efficiency, but you've got thousands of those devices in the network.

(35:21):
And if you're thinking about that spend strategically and working with the right partners, probably partners that you've already got, and you've probably got people bang in the door trying to sell you the next thing in this space, but it's kind of lost in the dungeons of the company. It's not the most coolest stuff like AI that you're working on. But let me tell you those areas and those companies are leveraging AI today as well. They can have a massive benefit and then how those organizations work with your other vendors so that if we were bringing in a new MX platform or we're bringing in a server rack, how are you working with all of that ecosystem to ensure that you're doing that in the most efficient way? What I see is quite a lot of silo thinking in telcos, whereas in the hyperscaler space, that's one of the core areas of opportunity because you're putting in lots of cabling, lots of metal work and materials and don't forget that has an energy usage that's further down the line, but it's equally as important.

(36:28):
So ensuring that you're looking at that whole process from the concrete in your network center all the way up through the tin, the metal and every part of it, you need to really be looking at all of that. Otherwise you might be saving in one place only to be driving cost in another. And I don't see telcos and service providers taking that infrastructure spend as strategic as they should and every year they have to spend this money because there's new health and safety laws or there's things that just need replacing or there's new environmental laws, things like diesel container systems have to be continually updated. And if you can update that and build a more greener or more sustainable plan, that can have a massive impact, especially if you're working and as Merkel said, Colts organizations got thousands of sites all over the globe as you're going to be continually investing in them, then think of that core infrastructure spend much more strategically than you do today.

Ray Le Maistre, TelecomTV (37:29):
Okay, thanks Neil. End to end thinking that's what we're talking about here. Now we're coming towards the end of our discussion here, but we've got one more question to wrap up and that is for network operators looking to use any AI tools to improve their network energy efficiency, where should they start? Beth, let's come to you first.

Beth Cohen, Luth Computer Specialists (37:55):
Oh my god, where should they start? Well, I think that the low hanging fruit is focusing on not necessarily ai, focusing on automation configuration, as Neil mentioned, turning on the energy efficiency configurations on your hardware. And that's pretty simple, right? Going through your network and figuring out what should be turned off. I had to laugh when I heard that story because years ago I ran a lab and I had to move the lab and I was turning off all the equipment and there was this one box that didn't have a label on it and I didn't know what it was. And so my staff was like, oh, what should we do? And I'm like, well, we have to turn it off. Well, I found out what it was, it was the DNS one of the top level DNS servers for the world. And so it quickly got turned back on again in a different location with a label on it. So just knowing what you have in your network and optimizing the equipment to make sure that it does run as efficiently as possible. And of course obviously negotiating with your energy, your providers is super important. And none of these have anything to do with ai,

Ray Le Maistre, TelecomTV (39:29):
But all good points. So to come back to Neil's point there, Beth, this end-to-end thinking needs to include label management. That's a lesson to take away from your little anecdote there. So Neil, let's come to you. Where should telco start when they're thinking about using AI to improve their network energy efficiency?

Neil McRae, Juniper Networks (39:53):
Yeah, I think looking at the overall architecture network, how can I take layers out? We've had this optical layer in the network for a long time that I'm just not convinced drives huge value, and I think we can do a better job at the routing layer and we see many network operators heading in that direction. And actually it sounds really hard, but actually it's relatively simple because in most networks you've got an element of redundancy and you can belayer quite simply. I think that's definitely one place to start. I think we've got, in our platform, we've got a tool called Paragon that can help you plan that. I think the second thing I would say is your own engineering expertise. People are passionate about sustainability, particularly engineers and because they know that if we save some money on energy, I can invest that money on perhaps another service or some other capability in their organization.

(40:53):
And I think there's so much data in the engineer's head that with a little bit of extra bandwidth, they can make a big difference in sustainability and they want to make that difference. Ray, we see a lot of our customers being really proud about the fact that they have saved or they're at the top of the league table for having reduced energy year over year and it really brings a added engagement benefit from your engineering teams who want to be part of this and who've probably got most of the knowledge that you need. Not all of it, but quite a lot of it. And then the last part I'll mention again end. We have a customer here in the UK that's using Google Maps to preplan the last mile build of fiber and that isn't an electricity energy saving, but it's an energy saving in that when they turn up to do a job to build that last mile of fiber, that they get it right first time as opposed to turning up and realizing they haven't got the right vehicles, they haven't got the right safety requirements or whatever it is just by thinking about your whole day-to-day process and how that affects energy usage both from the grid but also in diesel or transportation costs, that can also be a big factor and also as a vendor.

(42:20):
So Beth made an interesting challenge about this not being top priority for vendors. I think what's not being top priority for vendors is telling about what a great job we're doing in this space. And I'll give you one example. The thickness of the metal in one of our devices can be a differentiator between how green a device is and how unre it can be. And we put a huge amount of effort into every aspect from the thickness of the metal to how we mount it in cabinets to the packaging to ensure that all of that is as sustainable as it can be. And actually I think we've done a bad job of sharing that information with the wall because it's one of the things when I joined Juniper a couple of years ago, one of the things I was most impressed about was just we had this great story but it wasn't really well understood. For sure, we've got more to do and we continue to push that, but we want to work with our customers to see how we continue to drive those optimizations. But those are some of the places that I would be starting with and to reemphasize one of them, your own engineering team because they've got the knowledge in their heads for sure.

Ray Le Maistre, TelecomTV (43:35):
Yeah, well, I mean you've always advocated talking to customers, talking to staff, and there's obviously a lot of gain to be had from that and we could all use some positive stories from the industry as well and use cases. So we would encourage all companies to share what they're doing to make a difference. Mirko, let's come to you and then we'll end with Diego.

Mirko Voltolini, Colt Technology (43:58):
Seems to be the answer to start with is actually don't use ai. The basic has been said. I think if you look at the efficiency you can get with modernizing the network is amazing. We have been able to reduce energy consumption of certain layers of the network at the routing layer at the I layer by over 90%, introducing coherent optics and refreshing the technology with the newer generation technology, which is heavily energy efficient, I don't think usually should embrace using AI unless you have done the basics because you're going to try and optimize on a non-efficient network. Having done that, I think you need to first look at the use cases. What are the things that could bring more efficiency? There is no, I think from our experience, there is no tool you can buy today that can help you specifically optimize energy consumption based based on ai. So a lot of this, what we have been doing is actually internal builder. So you have to think about how you're going to create the capability to enable your company and this is actually agile development to create these use cases for ai.

Ray Le Maistre, TelecomTV (45:26):
Okay, thanks Mirko and Diego, final word to you. Where should operators start when they're thinking about how or if to use AI to improve their energy efficiency?

Diego Lopez, Telefonica (45:40):
Where they would add towards, my colleagues have said this, to use data, you have to understand where your data sources is are. You have to understand where your data pre-processing an identification should be located. You have to generate massive amounts of data of of many different kinds. I'm not talking only about the raw data that you can use for feeding an AI or training the ai. I'm talking as well about data's knowledge, these golden nuggets that Neil was mentioning at the beginning in the brains of many engineers using at the seed of what you start to do and then something that is important once you have in place those data. Something that is important as well is that you take into account that normally you would need data, unusual data to train the systems to validate the AI systems, to be sure that the AI system is going to behave appropriately in unusual cases. And for this, you need mechanisms for either share the data with some of your fellow participants in the telco industry or generate them by means of what we were mentioning before, digital twins or synthetic environments in which you can experiment. But whatever the case AI is based on data, ai, AI consumes and produce data, massive amounts of data and we need to start with these accurate and trustworthy data.

Ray Le Maistre, TelecomTV (47:27):
Okay, great insights there and clearly one of the key takeaways from today is that it's really important that network operators mine not only the data they have in their network, but also the data and information and expertise that they have in their engineers. But we must leave it there. I'm sure we'll continue this debate during our live q and a show later. For now, thank you all for taking part in our discussion. And if you're watching this on day one of the Green Network Summit, then please send us your questions and we'll answer them in our live q and a show, which starts at 4:00 PM UK time. I'm sure there are many questions we still need to address, so let us know what you want to hear. Discussed now the full schedule of programs and speakers can be found on the telecom TV website and that's where you will also find the q and a form and our poll question for this summit. For now though, thank you for watching and goodbye.