Cool. Alright. Well, thanks to everyone that's joined the webinar this afternoon. We'll be getting underway, as of now. And my my name my name is Joe Kershaw, and I am the channel development lead here at IP Fabric.

Bit of a nondescript title, but I focus on commercial relationships with some of our global partners and get involved with a number of our strategic as well. This is across media, but also across other, verticals across the market. And I'll be bringing some insights from some of our media and broadcast customers and partners that lead in this space from the US market to the more mature adoption markets in the in Europe and into some more conservative markets as well, looking at organizations across national television, across, some of the top 5 streaming services globally that have adopted assurance, that have adopted network assurance, particularly from IP fabric and applying this in their, in their approach to network automation. The technology itself does span across verticals, but what we're seeing that made us pull this webinar together is the new trends in media consumption, or maybe not so new, but over the last 3, 5 years, new trends, new demands are driving greater innovation in the media space and in the broadcast space. And we're seeing specifically for network teams, infrastructure management teams, that development, automation, innovation in general is is is a prerequisite to being able to compete in this market.

So we're seeing some really interesting use cases. Wanted to bring you our view from this market and then also share you a little bit about what network automation means in this market, and particular what IP Fabric can bring to, to bear in this, in this arena. Darren, do you wanna quickly introduce? Sure. Yeah.

So, my name is Darren Fulwell. I'm a product evangelist, with IP Fabric. My role, I suppose, is a little more technical. I work with customers generally on understanding, what their use cases are gonna be for for network assurance and how we can bring IP fabric to bear in their network operations to help and augment existing process, but also create new ones. The main area, I suppose, with that is network automation as Joe's already touched on.

And and I suppose for me as a as an old school network engineer who's, you know, showing the scars of of, many years of doing this stuff, Network automation, I suppose, has really taught me that old dogs can learn new tricks. Right? This is this is the way that we need to we needed to do things had to change. And so so here we are, in in 2021 where network automation becomes a thing. And it's a proof of how things, I suppose can change is is similar to the the broadcast media industry.

When I was a kid, right, this was a sight you saw often, when you sat in front of the TV. I was I was a child of the 70s 80s and in those days there were well there were 3 TV channels, believe it or not, and a screaming cat in the background as well. With where I suppose the the point is that there was no options. Right? You were you were you had time limited, viewing.

You you woke up in the morning and the tv didn't come on till sort of 10 in the morning or whatever you'd go to bed in the evening and the tv had shut off at 8 o'clock 9 o'clock in the evening Everything was pre recorded and there was no interaction. It was it was a very static medium just like the radio but but with pictures. And of course the quality check the fuzzy quality on there I think that's that's amazing but I suppose go on Joe. Yeah it was it wasn't just specific to your generation, was it? As a nineties kid, even when Sky came into the play and started buying up football rights, putting out new services that you'd never seen before, it was still a case of you watch what is on.

Yeah. And if there's not something on, you just don't watch. It's not the same sort of thing that we're looking at now, but even then, service disruption, no no programming being shown in the early nineties. It was exactly the same story. Yeah.

And and, of course, that service interruption thing, hence the test card was was regular thing. Right? You saw it all the time. Much of that was like most things in the seventies eighties, I suppose, but that's another story. But but some things don't change.

Right? We still get service interruptions, and we still see problems. This this is a great example. Whole TV channels going off air because of technical issues. Once upon a time, that was just an inconvenience thing.

It was like, oh, I wanted to watch that. But now, of course, you know, things have moved on. This means a whole load more. Right? With Absolutely.

This happening just 2 weeks ago, obviously, for the national television channels, it's a different thing. It's not quite the same competitive market where they're really competing over and over. But the idea is they only get the advertising based on their viewer numbers. Their viewing numbers can only be maintained with a with a concurrent service, high service value, content being delivered all the time. And we've seen this just a couple of weeks back.

And and what does it really mean for the networks? I've not looked into the pure detail of what this technical issue is that brought the channels down, but we know for the networks and the infrastructure teams that this situation will have alarms organization. And it's a case of if it is the network, how quickly can we get to a root cause? How quickly we can we validate what's potentially being changed or what's impacting it and get to a solution? If it's not the network, how quickly can you get to that meantime to innocence and move away from wasting resources in the area that's not to blame?

No. Absolutely. Yeah. And and what we're actually seeing now once we can start getting away from the the national television space, but also the the national TV programs are having to compete in this space is looking much more into a different way of consumption. And the idea of multichannel, multidevice interactive viewing with the ability to pull up in race data through the Eurosport app whilst you're watching on your TV, the actual race as it goes on, where in studio, presenters are able to pull up elements of the race for in-depth analysis.

The idea here is is a complete immersive experience, getting users hooked on the experience, not just on the content. And if you see, obviously, Eurosport competing with Sky compared competing with BT Sport and everything else within the market, It's about full user experience, full immersive based experience when viewing, and this is where the competition really kicks in. The only way that these types of organizations can compete, can fight for loyalty in a market where loyalty is at an all time low is by producing content element after content element and producing these experiences where their applications can hook into the viewing experience. Yeah. And I guess as well, you're you're not just competing with with traditional, media, providers either.

Right? You've got you've got apps on phones. You've got you've got YouTube. You've got TikTok. You've got all of these other things that are there are grabbing attention and trying to steal away, your your your, viewers from your your immersive experience.

So you need to do something different. You're just gonna lose lose those people and and lose the, the hooks into into their day to day lives. Right? I think that's that's the critical point. The the idea of consumption of the media, of subscriptions, of loyalty, this is all business talk.

It's not specifically networking talk. But if you work in an organization where the senior executives are obsessed with the idea of business transformation, they're obsessed with the idea of more content, faster, more services, faster. The idea is it's this ever increasing scale of expectation. But even working with our customers where we're working, as I mentioned, with some of the the well, the top five streaming organizations globally. We're working with Major League Baseball, obviously looking at, sports, broadcast live feeds and all of the elements that come with that.

These organizations are obsessed with further development, but they're not flooding their teams with more and more resources. So so how do you reckon it's possible that this ever increasing spike of expectation is managed with exactly the same number of resources? Well, well, you know, we've already we've already touched on on network automation as being, an important part of this. And I think network automation within a larger framework of of, operational framework is key to it. Right?

It's about being able to do the same or more with less or or or, being more efficient about how you approach things. Now I know we've we've talked about this before and we we came up with a few, a few areas where things are changing in, in the technologies that are being used to deliver these kinds of mechanisms. Right? And these are these are some of them. There are more, of course, but these are 3 that we picked up particularly.

In in the first, what we're talking about here is the use of of IP networks to, I guess, manage the the the capture of video, the transport of that video in order to then be processed, and then the distribution out to a customer. So that the idea of being able to manage it as as IP data from end to end, from from producer to consumer, which in itself means a whole load of chunks of different network working together in order to deliver that. Right? You've got a you've got some sort of access network where you're where you're bringing those data streams in. You've got to then get them into some sort of facility where they can be processed and brought together, and then they're streamed out to wherever the customer is located.

And so there's wide area connectivity, all kinds of things, involved there. So there's there's a lot to to manage end to end in order to deliver that that, capability. And and go on, Jeff. You're gonna Yeah. I'm just thinking and would you would you approach this kind of balance between trying to get the unlimited lake of data access that the cloud could potentially offer and trying to balance the amount that you would be adopting because of its impacts on bandwidth latency.

Right. That that balance is only one element within this complex puzzle and automation surely can't be the answer to all of this in like step by step pockets of isolation. This is this. So this is the the problem that that you're trying to deal with because network automation, the way that we've adopted it so far has been very much focused on individual, individual domains of network. And you have to get that oversight in order to to really get the best from this.

You've mentioned the cloud adoption stuff here. Right? So cloud's great for for for piling storage, being able to to basically have bottomless pits of storage where you can just put all of the data to process, but there's a latency involved in getting stuff to and fro. And there's and there's a way you need to manage the security in the cloud and and the the connectivity that people aren't familiar with. So you've got additional tooling required or you've got new skills that need to be learned in order to deliver that.

You know, so so you're absolutely right. It's it's about being able to bring all of the different elements together to work in concert. And this talks to, with from a from an actual configuration perspective, it talks to orchestration and that sort of thing. But this is where Assurant allows you to get the visibility of all of those those areas, bring them together into one consolidated view of the whole environment to understand the impact of each of those different elements on that end to end, service. And I think if you but just before we step into to what Assurant really is, then again, into the the nuts and bolts of it, although don't fear everyone, we're gonna keep any demonstration pretty high level.

I think the the idea we wanna get to is is exactly that thing. The executives, the business, the product management, the content management are all looking at more, faster, better, and and trying to capture that elusive loyalty from customers. Whereas the network engineers today are often looking at networks, devices, pieces, elements, WAN, LAN, DC, campus, whatever they're looking at, they're looking at it in elements. And so there's, there's almost this back to back trying to tackle the same challenge as opposed to both looking at it from, what service are we delivering? How is the service built?

What would changes mean? What would the dependencies of these changes mean to other areas. So how can we use network assurance to change the way the network engineers to look at the full picture, the end to end connectivity, the service delivery, and maintain it as a continuous service without any disruption. I think that's it. In a more nuanced sense, it's about changing the way that teams can collaborate with each other in the technology space.

I was just gonna say, so because again, what you're talking about there is a is is literally different groups of people. Never mind different technology domains here. You're talking about, you know, you're talking about a cloud team supporting, a network team or a security team or whoever. All of these different people with different perspectives on the same problem, of delivering that service. And what you really need is that that single pool of data that they can all refer to that let's see.

You get the consistency of of the viewpoint. But, also, you it allows the individuals to see beyond just the for just their silo. And I think, you know, from an operational perspective, we we you know, you hear about breaking down silos as being a thing that we need to do all the time. The only way you're gonna do that is to have the the the right data, the right processes built around that data in order to deliver that. Right?

Yeah. Absolutely. So what we're getting into here then is specifically what network assurance is and what we've done with these points, which we're gonna feed into into a very brief demonstration, is we've kept them as higher level capability with the challenges in mind, with the idea of the thirst for innovation, yet the balance with service consistency, with deployment consistencies across teams, security policy consistency, looking at Assurant more as a capability for an organization, and then we'll bring up a couple of demo cases that can show exactly what we mean by these. But would you would you add anything, Darren, around what it means dynamic accurate data for whoever needs it when they need it? Or you can say self explanatory?

No. That's absolutely of course, that's gotta be the case. And and and I suppose the beauty of of the way that IP fabric, delivers that is that that that isn't just a one shot thing. It's not just a baseline. You get that initial view, but then that data is refreshed, constantly with that end to end view every single time.

And and every element of of what we'll look at is updated each time that snapshot's created, each time we go through and and refresh that data. So, I mean, for I'll give a quick example, if I may, Joe. I'll just, I'll just drop my, my screen in here. We'll go for that one. Hoping you're seeing the right thing.

Yeah. Yeah. Swindered. So this is, obviously, this is IP Fabric, and and it's GUI, but and just showing some inventory data here. But the sorts of things I'm talking about, I suppose, there is this this ability to track, change, between between different times, when when snapshot is created.

In this case, we've got a snapshot and a snapshot taken the day after. And what I'm doing here is just basically determining which devices have been added and removed. Very simple, but it's accurate data that's been pulled from the network across all of the domains. In this case, you can see a new location has been added into the network. And so a bunch of devices have been added.

I I think just just an interesting point on this. We sometimes get asked by customers who have seen every manner of tool in the past about, yeah, well, we've we've got this data. We've seen this data in our systems. Why would we look to change the way we manage data in our network? And I think the point with this is once you start getting towards automation, there's the whole nuanced conversation around source of truth.

And without going to elaborate too much on source of truth, the difference with IP Fabric, and it was the 2nd bullet point that we just pulled up without merging them too much, is this proactive analysis. But the way we're reading it is the source of truth is actually your network. An IP fabric is a lens through which you're viewing that source of truth from the network. We're taking exactly what is discovered in your environment and being able to allow you to model that over time to be able to make these type of comparisons. No.

It's a good point. I mean, the point, I suppose, when you're dealing with with an automation platform, for example. So and, again, it's it's a nuanced thing, but but source of truth actually is the source of intended truth you're talking about. Right? It's that that thing of being able to say, this is what I want my network automation to do.

What we're looking at with IP Fabric is that source of observed truth. Right? That that this is what the network actually is, and this is how we've interpreted it and how we built a model around it in order to almost to see the difference to see what what what needs to be done in order to bring the network closer to its to its desired state, which is yes. It's it's I know which devices I need to be in and which they're connected to, but it's much more than just that. Right?

We've we've already talked about this before, but if I, go on. Yeah. And, yeah, if I mean, it just feeds into this use case. If you've then got an operational team looking at tactical data for changes that have been requested, But then you also have an automation team that are looking at the overall view of changes being made in the direction of an intended stay. How would you get both teams or both professionals operating from this same set of data?

Well, for 1, you can just give them access to the platform, of course. But but other ways you might do that is to find ways of pulling this data out. And and it's really, really simple, obviously, in IP Fabric. We have from, every single table in the platform, we have this ability to to act access the API documentation that extracts that that very data that you're looking at here. Now that in itself is great because it means you can write scripts.

You can you can build automation to to go do that stuff. But you can take that a step further and feed that into chat ops platforms and and this and the like where this stuff is proactively pushed to people so that they get notification that things have changed when IP fabric discovers them. And I think that that's where it gets really interesting because you're you're then proactively sharing that data with whoever wants to use it. And it it provided that sort of curated view if you like of of the of the data that's there. But the source of the data is the same regardless of whether you're using the UI or whether you're using other systems that are integrated with IP fabric over API.

Does that make sense? Yeah. Yeah. That's just good. And if we were to to kind of move into that second bullet point, then proactive analysis and how we're capturing this data on the regular basis, What does that mean for for service consistency?

And if we touch on the word compliance Yeah. How do we tie that to service consistency and and consistency of configuration? I mean, this is this is the thing. Right? So so it's and it's not just configuration that we're extracting, but but that state data as well.

So we understand the how things are configured, but we also understand how they're behaving at that point in time. So we extracted all of that. And what we're doing is we're pulling this data into tables in the platform, and you can see there's a whole bunch of of technologies that I've got listed out in this menu here. I'm just gonna pick, one which is a very typical one, where you're running BGP routing, for example, in your network. We have all of this data available to us.

It it changes, from from, snapshot to snapshot based on the state at given point in time. And what we do is we color code it. So you see all these colors, looks a bit garish actually, but, but there's a reason for those colors. They're there to to indicate to us where things are compliant or otherwise with state that we're intended. Okay?

So we have those intent verification rules that we list at the top there. And I'll just pick 1, for example, the the nay BGP neighbor state. This this state column shows us whether the neighbor relationship is established, whether it's, in transition or whether it's active, I e, it's not yet been established and there seems to be a problem with it with it doing so. We capture that every single time we run a snapshot, so we know what state that's in. And that data is available to us in a dashboard and reporting it again through the API if we want to.

So we can we can bring that data out and place it in a put it in other places. But we can filter the tables based on that. So if I want to know where there are all the the active, BGP neighbor, where where things are trying to establish but not able to because perhaps 1 only one end is configured. I can identify that list immediately and either provide that to whoever needs to go fix it or we can, perhaps raise it, a ServiceNow ticket for the back of that with that information to say go go fix these things. The point is what you're doing is you're building that the and that they're behaving the way you expect them to.

And those those two things are are really key. K. And, to to take you back to the use case there of then applying this to service delivery and broadcasting, You get isolated projects run within network engineering that specifically go off and do an audit for BGP. And it's an engineer, probably a well paid engineer that sits there doing work he's really not happy doing for multiple weeks. In order to pull back a report like this, this effectively saying, yes.

We've got problems. Once you start looking at actual service delivery for broadcast, you're looking at content delivery networks, you're looking at consistency of service. Those multiple weeks of audit time could have been spent in fixing time and could have been spent in service response, ticket resolution, any other things. Would you tie anything else to to get a service consistency and delivery there? I mean, the other aspect of that is, of course, the data that you're you're delivering as part of an incident response or whatever.

The actual troubleshooting. This this is something I mean, I was talking about, this to someone the other day. Effectively, what you've got here as well as all of this proactivity and everything is is an underlying, documentation platform with all of the data that you might want in order to validate that that the network is doing as it should. Now when's why is that important? Well, if you're troubleshooting a problem and you go to your documentation store, how do you know when that documentation was updated?

It might have been when the when the the network was originally built and not been updated since. So it may have been updated a couple of changes in, but not not necessarily accurately. So what you're having to do is each time you pick that documentation up, you're having to validate that it is actually correct before you can even start to do whatever troubleshooting you've got to do. Or you may not have the right detail in there, for example. So with IP fabrics, Assurant, what you've got is a a a view of that, of that documentation, which you know is up to date, which you can trust, has all the data in it that you need.

And you can just get straight to work because you've got all the information to hand. You don't have to go building your own. You don't have to go mapping things out with a pen and paper because it's all here for you. When you touch on mapping things out, pen and paper based, when you talk about incident response, where you talk about these sorts of situations, the 3rd bullet point, which there's no point in me resharing my screen, was service level mapping for operations to be able to look at network resiliency in these type of situations. So if you're talking about pen and papering a service line within a broadcast network, what what would that look like?

Well, you've got you've got, our our ability to perform an interim path simulation across the network. Right? So so we can take a a view simulation across the network. Right? So so we can take a a view of and I'll and I'll show you a very I'll show you a couple actually.

So this would be like a unicast, conversation from one end of the network to the other. You've got an endpoint trying to talk to an endpoint and all of the devices that you're gonna talk to all the way through. So this is a really simple version of, what a what a packet flow might look like through switches, through routers, through firewalls in order to get to to an endpoint. But we've got, we can take that a step further, you know, something that's very prevalent in in, media distribution in particular is multicast. We have this ability to do the similar a similar thing, but with a multicast tree.

So in this case, this is a distribution tree from a particular host, and I'll just get rid of that and make it a little bigger, in order to from from a host out to all of the receivers who've registered to receive that Montecast, stream. Here, you can see all of the steps that's gone through, all of the routers that are involved in the process. And, I mean, just to to clarify, I mean, if I just show you a couple of these, this router here is a is an Arista, device. You've got, let's see what this one is. I think that one's a, Juniper there, and you've got, a Cisco, if I'm not mistaken.

Yeah. There you go. Cisco device there. So you've got a whole bunch of of devices from different vendors, all contributing towards delivering this service. And what IP fabric's able to do because of that path simulation capability is it can it can track through all of the devices that are impacted and highlight them, very simply, here.

But also, we can dig into the actual path that traffic's taking. So this through the path inspector here, you'll see that we can check to see exactly what's happening, what decisions have been made through the path, even down to the level of if you're that worried, what the packet headers look like as you're passing through from one device to another. It's you know, this is a a 1000000 miles an hour view of this capability, but, obviously, something that's that's hugely useful because otherwise, I'd have to get someone to sit down and map that out by hand. And and if they don't know Arista or they don't know Juniper or they don't know Cisco, they're gonna struggle to understand that full path. I mean, to be honest with you, multicast routing is a bit of a dark art for most people anyway.

So And if something within that path has been changed in a device, how is that engineer supposed to to map that out from device to device and understand the impacted point compared to to when that service was looking good? Yeah. That's a really good question. So we've we've got this ability, to overlay. So the compliance information that we we showed before in the color coding, we can actually overlay that information directly, onto the map here.

Now I'm looking we don't actually have any error conditions in the multicast here. But what I can show, for example, is that everything is good. So if I if I show click on this one here, which shows shows me PIM session age, for example, this is the the protocol that's used in order to distribute multicast routes. You can see that everything is green. Fantastic.

If there was an issue with any of that, it would show up red, for example. So so we're able to to understand the the detail of the protocols and the detail of what's going on under the hood and show it visually by overlaying it onto the, onto the path simulation. Nice. So if we, if we just bring it back to, to summarize and to, to wrap things up, Obviously, the capability we've just spoken about is not network automation in its own right. I know you've recently, been heavily involved in the white paper that's that's been released.

Do you want to add a little bit more color of what Assurant is within network automation and and how people can get hold of a bit more of that information. Yeah. Of course. I mean I mean, yeah, we've we've looked at it very much here as a as as a an interactive tool. Right?

IP fabric. We we've seen here how we can interact with the data that's all available and so on. I think the important part with from a network automation standpoint is that the availability of that data. Typically, if someone's building network automation projects, they're gonna have to go and fetch all of the the data that we've gathered here themselves in order to process it and deal with it and manage it. So so straight away, we've we've got one API that gives you access to all of this data including even these images that we're seeing in front of us here.

We can we can extract those through the API as well. So so that's the first thing. I guess the the second element to that might be, the the approach to actually making changes. So because we have this capability, what we can do is we can we can visualize what the, and and understand what the network looks like before a change is made, execute that change, and then create a new snapshot to show what the situation's like post change. And again, because that data is available through an API, we can make a decision outside the platform to to back out that change if needs be.

So you can build it into your pipeline for your change of make change, pull new snapshot, check that the problem has been fixed by that change or expected, state is as we as as we hoped it would be before the change. And if not, then back that change out or execute some other process. And so you're building you're building, on the data to to make decisions in an automated way that you don't need to be involved with as an individual. That's that's the point here. Right?

You're making this data available for these purposes. And I guess the the other element, is the fact that we can we can both trigger, we can trigger activities from the data that we see in IP fabric. So when we run a snapshot and we create these, these compliance checks, IP fabric can send out triggers to other systems to say, go do something with this. Go make something happen. And we and so while we don't actually automate anything in IP fabric itself, it doesn't do any config push or policy push.

What we can do is use it as a trigger to go ahead and execute some other process in order to to to deliver that data into an automation platform. Brilliant. Alright. Well, if you've if you've not got any little gems to add at the end there, Darren, I guess we can, we can wrap it up for today. I think I'm good for the moment unless we have any questions.

Just looking at the panel. I don't see any questions in the, the q and a panel at the moment. No. No. No.

Okay. Okay. Well, thank you everyone for for joining the session this afternoon. We will be posting out the recording, and we'll also share the links with that recording. So you can look into a bit more of the information, including Darren's white paper recently on the road to self driving networks.

Thanks, Darren. No. It's a pleasure, Joe. I'm looking forward to the next one. Speak soon.