Cloud OnRamp Unlocked: Seamless SaaS & Multicloud Access

Hello everyone and welcome to today's expert insights session. Today's topic will be covering cloud on ramp, seamless SaaS and multi cloud access. So over the next hour.00:09More we're going to take a deeper look at how Cloud on ramp works in a Cisco Catalyst SD WAN environment with a focus on how you can use cloud on ramp features to improve user experience in the network and seamlessly connect your network connect your network to the cloud. The goal of the session is not just to show you where things are in the dashboard, but to also help you understand.00:20What cloud ramp actually is and what options we have that we can use to, to bridge the gap between our on prem networking instances and the cloud itself. So we'll walk through concepts step by step. We'll look at some differences between.00:40Cloud on ramp for SaaS, cloud on ramp for multi cloud, and then we'll go through some configuration examples along the way. So with that, we'll go ahead and get started. Before we dive too deep, I wanna briefly show you one of the next upcoming EIS sessions that we have where we're going.00:56So it'll follow the same format as this one essentially or in our previous ones, but the next one will go, we'll dive deeper into migrating from templates to policy groups. So this is that UX one.O to UX two.o, you might have heard about whether it's from Cisco live or from your Cisco account team, but this session will go into some of what the best practices are.01:15From migrating into that new deployment model.01:35But definitely highly recommend attending that. You'll get some good value out of that one as well, and that's next Thursday, the 23rd. Our presenters today, we have myself Chad Ralph, and Edgar about Valvedia will both be presenting. So I'm doing the main content. Edgar will spend time going into demos following.01:39Each section.01:59A little bit of a little bit more about myself. I've been working in enterprise networking for, I guess over 20 years now. It gets it's kind of goes by in a blur, but of of those 20 i've spent the last five at Cisco in customer success primarily covering enterprise networking, so Catalyst Center as well as.02:01As SD WAN specifically. So this has been my role.02:20The entire time I'm at Cisco, so, but for the last five years, I am, I held multiple certifications, I guess specifically as it pertain to this topic. I am a CCIE enterprise infrastructure as well as I hold SD WAN certification specialization as well.02:24With me, we have Edgar. Edgar, if you want to go ahead and introduce yourself. Thanks Chad. Hello everyone. My name is. I also have been on Cisco for some time as you can see, more than 20 years in the experience in networking about seven years in in Cisco as customer success specialists and I also.02:42Work with other enterprise technologies like Catalyst Center.03:01I also spaces and obviously as the one that is the the the topic that we are now covering, right? So I also have my my backgrounds on certification and I have also CCIE and enterprise infrastructure but also SD one certified. So it's a pleasure to be here with you and also with Chad sharing this, this.03:06So go ahead chat chat, please. Thanks here. And along with us, we have Tushar and Eris who will be answering questions or our panelists for this session. So as you have questions, as they come up, definitely put them in the Q and A section of the Webex application and Eris and Tushar will spend some time in.03:26Answering those throughout the presentation. But yeah, definitely don't be shy, ask what questions pop into your mind and we'll definitely get you an answer as we go along. Eric Shar, thanks for your help with that as well and Edgard, definitely thank you for joining me in this cloud on ramp journey today.03:45The agenda, what we're gonna cover, we're gonna start with cloud on ramp use cases, right? It's a, it's a it's a fairly generic term. It covers an umbrella of multiple aspects of cloud on ramp features that we can deploy within the catalyst center or sorry, within the catalyst SD WAN environment. We're gonna then dive into cloud on ramp for SAP.04:02Specifically so cloud on ramp for software as a service type deployments, then we'll, we'll dive into a demo following that section. Then we'll come back to cloud on ramp for multi cloud, which is again one of those larger umbrellas that covers multiple aspects, but I think for for most, right, this is going to be the cloud on ramp for what you might have historically.04:22Sort of infrastructure as a service, so AWS, Azure et cetera, right? Those workloads in the cloud, but it also covers some additional features like cloud on ramp for interconnect or SDCI et cetera. And so we'll touch on all of those. We'll then go back to doing some demo in, in that environment to show you that and then we'll following that.04:41Demo. We'll do some wrap up and end the call. Following this presentation or the the meeting today, there is going to be a survey sent out but only to those that are still on at the end, so make sure you stay through the end, you'll get the chance to tell us how we did with the with the presentation, the, the topic, the.05:02Actual components, right? Well, what you learned, what was, what can be improved upon etc. So definitely love to hear your feedback. So stayed at the end and do that survey if you can, we'd love to hear how we can improve.05:22So again, starting with our cloud on ramp use cases, I'm gonna start with essentially, you know, what problem are we trying to solve? And I think it's one that we're all, I mean, more than aware of, right? As long as if you've been in networking for some time, the one constant is the networking networks in general evolve. It has evolved, right? And a cloud.05:36Isn't anything new, but as we've evolved, things just become more complex. We end up wearing more hats etc. I mean applications are NO longer only in our on prem data centers where we control the lion's share of the path, right? We have both ends so we can, we can make forwarding decisions relatively simple since we own that infrastructure.05:55Now we've got to deal with applications and workloads hosted in multiple data centers as well as in multiple cloud providers, right? It's NO longer just simply.06:15Okay, I have a data center and I'm backhauling to AWS like it was in right 2008 or whatever, right? So now we've got applications hosted in multip multiple data centers and then directly over the internet with SaaS applications as well, right? Those are, they play a very large role in a lot of enterprise networking, right? Whether it's office 3605.06:25Five or voice services in a lot of cases are hosted over the internet, you know, box, SharePoint, whatever it might be, right? That's another path that we have to be, that we're essentially responsible from the user's point of view to deliver some, some, good user experience generally, right? So how do we.06:45Optimally connect to those services while still providing the service that our customers are expecting or in a in this case our customers, meaning the users in our, our network, right? Cause they don't really differentiate necessarily between right to them, it's the internet is slow, right? Or the network is slow if it's things aren't performing, so we need a way to.07:05Provide that secure access so that's that's always critical and any deployment has to be secured but also in a way that our customers at least get the same level of, of experience that they're used to in their environment.07:25And how we address that is with Catalyst SD WAN cloud on ramp, right? So obviously catalyst SD WAN accomplishes more than just the cloud aspect of things, but for the sake of the topic today, we're really talking about how we address those diverse application loads in multiple locations, right? Because those multiple domains.07:40That we're responsible for, and within the cloud on ramp umbrella, right, we're gonna talk about cloud on ramp for SaaS today specifically, so being able to use real time analytics for each application to steer users onto the best performing path for those applications that are hosted out out on the internet. And then we'll also.08:00Talk about cloud on ramp for umbrella, specifically talking about.08:20That infrastructure as a service connectivity, how do we connect to our public cloud or private cloud, right? So how are we connecting and and seamlessly, whether it's, you know, using a workflow to connect to that cloud or using something like cloud interconnect where we're automating on demand connectivity to multiple cloud services from some.08:24Place like an Equinix or or Megaport et cetera.08:44So we're going to 1st dive into our 1st use case, which is cloud on ramp for SaaS. With with cloud on ramp for SaaS, right? The, the, the plane connecting.08:50To a SaaS application has never really been the problem in most cases, right? If we have an internet connection, we have potential connectivity to the cloud, but the problem is that we don't control both ends of the transport in that, in that instance, right? So essentially it's kind of like fire and forget. We send traffic to the internet if it needs to go to an internet base.09:05Based application, and since we don't control the internet, right, when we send it out, we are just anticipating that we're going to receive it back and it's going to perform ok, right? We can't use the same tools in our toolbox to necessarily influence traffic once it exits our network. So how are we, how are we able to determine a best path, right? And.09:25Traditional networking or you know circuits between sites are essentially a bookend solution, right? We have control over both ends of the solution. That's the bookend analogy, right? So we can influence and steer traffic any way we like when we control both ends of the path, but when it's a SaaS application, right, we don't.09:45Right, in most cases, I know I don't have any access to a router sitting out in Microsoft's data center or, you know, whatever, wherever it might be to influence that path. So how can we possibly steer traffic?10:05According to best practice or best performing tunnels in that instance, right? So that's really the problem that we're trying to solve with cloud on ramp for SaaS, and again SaaS being software as a service specifically. So how can we improve, right? We've got multiple internet circuits. How can we determine the best path other than just right, I mean the one knob that we generally have.10:19Have to steer with is ok my local circuit is saturated, so use the other one. But that isn't telling us the entire story from site to SAS, right? So we don't know along the path that there's brownout type type issues in the environment. And when we're talking about the internet, there will be those situations, right? Where we have a port.10:39Performing circuit that we have or or a router or a peering point within the internet that we have NO control over and what do we do about that? Versus just say, ok, try it again, right?10:59So with Cloud on ramp for SAS, that's essentially how this is the tool that we're going to use to answer that, right? So we do that with essentially capturing DNS requests and probing those different paths. And so let's walk through a little bit on.11:09What that looks like, right? And so, starting with a user that needs to access or wants to access a SaaS application, right? We start with a DNS request, right? User goes to box or whatever application that we're talking about and 1st they send that DNS request. That request is intercepted by one.11:28Of our edge routers at the branch or their edge router at their branch, even though the client has a different may have a different DNS server configured or the router at that branch is going to intercept that traffic and then duplicate that DNS query.11:48And reoriginated sending out all of its available paths. That's step one, right? We are then going to do some, some probing along that path, right? We're gonna send HTTP or HTTPS ping packets or probe packets along each path in order to figure out which is the best, right? This is essentially how.12:02How, right? We we don't have that other router, that other bookend in this situation, so this is how we are going to determine what path is better.12:22In in this case, right? We're sending probes amongst our available paths to determine which path and essentially measure the quality of each path using delay and loss in this case. Based on that data that we receive, we calculate a, a quality of experience score essentially, and that score is what we're using to determine.12:32Path A or ISP one in this case is better than ISP two, right? So this is our determination for.12:52How we're going to pick one path versus another path, right? It's, it's really all about those probes. But before we get into those, it's take another view of that that DNS view, right? So if we look at it from a DNS perspective, what's happening in the environment, right? So in this case, number one, again.12:59Start with the DNS request to the SAS application which the router intercepts. The router then uses that application inspection engine, right? So we aren't using just DNS to determine.13:19The application that you're trying to go to. I mean that is part of it, but we also have a full DPI engine within the environment where we can validate that this is indeed the application that we expect it to be and and can treat it accordingly. So, and also we have, and we'll talk about that in some later slides, the.13:31The ability to do custom.13:51A cloud for SaaS applications as well. So we need that this step to to validate, ok, we've intercepted DNS, but we also need to compare it with our, our, our DPI engine or SAIE engine right to to validate that this is the correct application. We then overwrite the destination of the DNS request sending an out.13:54Out available paths to or with the the router that's or the DNS server that's configured on the router itself, right? So the router is going to intercept it and then I'm going to send the DNS request to the DNS server that I have configured in my VPN zero or my WAN transport essentially a VPN. I'm gonna send that out, right? And then once.14:13As I get the response, I'm gonna then update the destination, right? Cause that response came to me because I sourced that to my DNS server. I'm gonna then, of course.14:33Ultimately we do need to get that DNS response back to the user. So I'm gonna update the destination and send it on its way over to the user. So that's the the DNS perspective of this application flow. And then after DNS, that's when we probe, right? That's when we start those HTTPS or HTTP requests depending.14:43On the application, right? And so depending on your site layout.15:03Right, if you've got multiple potential exit points, you can, you would generally configure both of these. We're gonna measure and probe both of these paths to make sure that they are right. We have something to measure against. If it's only ever one path, then really this is kind of, kind of a mood point, right? But we need something to compare it to. So if we have a site as an example on the left hand side.15:07Like where we've got ISP one and ISP two, we're going to perform those probing on both paths and we're going to choose the best one of those. Another very common scenario for a lot of our customers, right? They might have a single DIA.15:27Types direct internet access circuit at their branch location, but then also like say an MPLS or a private circuit of some sort, whether that's metro ethernet, MPLS, whatever it might be, right? But in this case, this isn't direct to internet, but it's.15:42A tunnel to another site. Like for instance generally what we would call a hub site, but within cloud on ramp for us we would essentially call this as a gateway site. So essentially we can probe both the local circuit, right? So we're taking measurements locally to find ok to get to Office 3605.15:57This is the, the measurement. This is my QoE score for my local circuit, but I'm also then measuring both across my tunnel to my gateway site, so across MPLS or across essentially across the SD WAN fabric, right? We don't care that it's MPLS, but in this case, just whatever path that we need to get to what we would consider that gate.16:14Gateway site which again is generally whether it's a pop of your own making or usually most commonly this is one of our data center sites, right? Something that's already acting as a hub of sorts, right? But so we're measuring that path to that data center and then out from that site to.16:34Office 3605 or whichever SaaS app we have configured right for cloud on for SaaS. And so we are comparing these two, right? These are the two available paths that I have from this site. And there are going to be cases where we have some loss or latency.16:52Higher on the ISP one link and we'll send it across our SD WAN fabric out to a gateway site and out that way, right? And so we can have any combination of these setups in general, right? So we can cross and measure and probe on both.17:08Talked a bit about probing, but what is it I guess, right? So that, I mean, I don't want it to just be some nebulous term that, ok, don't worry, we send magic HTTP, but want to get into a little bit of the details of what that probe entails, right? When we're looking at it in general. And we start, I mean read the slide from the bottom up.17:24So we're starting with probes, we send one probe per second, whether it's HTTP or HTTPS, it's 1 /s, right? We're gonna do that. We're gonna send ten probes, so 10 s and then we're gonna sleep for 20 s, and then we're gonna do that again and again, right? And so essentially, right, these are measured in seconds, so we're gonna send ten.17:43And we're gonna sleep for 20, right? We're gonna do that again until we collect four sub buckets of 32nd probe intervals. That each equals.18:022 min, right? So for each average, so in order to calculate, we take a moving average of six buckets of 2 min each. So, right? Essentially the right we're sending probes out, we create our sub bucket and then all of our sub buckets together equal one probe.18:13Right, and so for every 2 min, we have a or sorry, not one probe, one bucket. And then we take a a moving average of six buckets. So it's a 12 min interval essentially is what this equates to, right? So we see that in minutes here. We're using that to determine for every 12 min, right? We're or it's a continue.18:31This moving, average of the of the those six buckets to determine the best path. So, so generally speaking, it, it may take up to 12 min to make a determination, right? On which path is better. And we do that, right? We don't we want to avoid.18:51A situation where we have flapping, right? One thing we don't want to happen is a constant failing back and forth where, ok, well, this path is better, right? Cause we're talking about things over the internet, number one, right? This isn't just over an MPLS link to our data center. This is to across, I mean, who knows how far.19:08Across the internet depending on which provider you have et cetera, to that SaaS provider. So we want to avoid situations where.19:28We're failing over every 30 s in that case, right? So, but, if our interface fails or if that local NAT exit goes down, we are of course going to fail over in that case and and to increase the speed of failover, there are other options as well, right? The, the 1st being using essentially a.19:35Tracker on the interface, so we can track something upstream, whether that's the actual SAS provider which would make sense or something else essentially, we create a tracker and track that. And if that tracker goes down.19:55Go ahead and essentially bring that or fail over that, that SAS application in that in that case. And the good news is with the config and policy groups, which is again going to be the session next Thursday, so perfect timing on this, right? If you are moving into that, when this is configured with policy groups and.20:07Later versions this is done or this will be the default, it's not done automatically, you still have to configure the tracker, but it is the default option, kind of new I mean that's in most cases something that you're going to want to use, right? In that case. So this is something other than alright, just the interface failing, but we want to track something upstream.20:28And with that probing, right, with that, that QoE score that I've talked about in the previous example I had right in that 1st section I had ISP one and ISP two for that dual DIA type site, right? This is just an example of, you know, what we might see from a scoring point of view. The VQ QoE scores are.20:47Quality of experience scores are listed above, right? And so generally speaking, right, it's not generally speaking, but it is actually literally one through ten is the score values, one through five being red or poor, five through eight being yellow or you know fair and the eight to ten being green. So, right, obviously in an instance.21:06It's like this, the closer to ten, the better it is, right? So we're, that's the determination we're making based on loss and latency.21:26And depending on the score, right, we're gonna make a decision and that score will always be one through ten. And this is just an example, right? We have the dual Dia site having the two scores ten versus eight versus the single DIA and gateway site or POP et cetera, right?21:34We do have certain instances or certain applications that we can take even a step further in our cloud on ramp for SaaS, right? You've seen already we can use cloud on ramp for SaaS to connect to some of our standard cloud on ramp applications, send HTTPS probes etc. But Cloud on ramp for Microsoft.21:52Office 3605 has some, some interesting additional.22:12Not just challenges, but it's also happens to be one of the most commonly used SaaS applications for enterprise enterprises. I, you know, I I don't know the specific metrics for it, but I know all of my customers, I don't think I have a single customer on my own that is not using Microsoft Office 3605.22:16In some form or another, right? One of the challenges though with Microsoft 3605 is that it isn't one app, right? So if you've got onedrive, you might compare that to box or teams or Office 3605 or Outlook or SharePoint, right? These are all individual tasks that fall underneath the Microsoft umbrella. So one of the challenges that we have.22:35With it is, hey, what if I don't use all of this? I don't want to turn on a bunch of probing test services that I don't use just under that umbrella and make a decision. I don't want to steer traffic if Teams is, if I don't use Teams and Teams is having an issue, I don't want to suddenly steer traffic to another link, right? If if Outlook is fine.22:55Right, if my email exchange is fine, I I don't necessarily wanna to to make determinations that way but others would be, we can categorize and get more dedicated telemetry from Microsoft to make more informed decisions, right? So with cloud on ramp for Microsoft Office 3605.23:14It allows us to add distinct URLs for applications where it has more information about what actually makes up those different applications versus just to make DNS specific a fully qualified domain name, right, in this case. So we can map those URLs to different service areas, and then we can categorize them however we want, right? We.23:35Have a little more flexibility with them to say, ok, I want to optimize this traffic or not in that case. And then with that, we can also enable something called Microsoft Informed Network routing, which what this is doing is this isn't just us, this is the other end of the book, bookend, right, in my previous example. This is allowing.23:54Us to get additional information from Microsoft so we we get direct telemetry from Microsoft on the health of certain applications and regions so that now we can make even a better and more informed decision based on information that they're sending us, right? So this is again.24:15A really good situation where it isn't just, I mean the HTTPS and HTTP probes are great, right? Making, being able to make a decision on a delay and loss over the internet is I think is outstanding, but being able to make a decision based on the company essentially telling you that, hey, I'm having a problem here.24:30Don't use this path is even better, right? Something that we don't necessarily have. It's not that easy to do across the board. So, with Microsoft informed routing, we get that additional telemetry that gives us that extra visibility into the application itself. So it's really, in my mind, it's kind of a twofold improvement. One, the most important.24:50Important improvement is we're getting better data to make a better decision so our end user experience is better, right? Ultimately, right, we're providing the service that our users are expecting and we're doing it in the best way possible. That's.25:10Job one, right? That's, I mean you can it's hard to ask for more than that, but we're also getting along with that some additional visibility due to that, that, partnership with Microsoft, right? That we don't necessarily have for those paths. So, within analytics, right, we can get some visibility into not only which path is good.25:24Versus bad, but we get some some history from that as well. So we have better data now just for our general service health and overview.25:44Then with the one of the other apps that we can do a very similar thing with, it doesn't have exactly the same issues but it is one of the services of course Cisco owns and and offers, but we have customers doing voice and video over Webex regularly, right? And so we can, of course, being Cisco to Cisco and we we have so.25:58Some additional flexibility there, but we can do a similar thing with our Webex, you know, cloud on ramp specifically for Webex, where we're not only just sending path monitoring probes across our links, but we're also getting better data back from the Webex servers with the Webex API integration, right? So this allows.26:18Tells us again like the Microsoft Office 3605 view.26:38We're able to make better, more informed decisions based on the data that we're getting from the Webex servers themselves, which gives us visibility into the actual network path via telemetry data that we're seeing from path analytics, right, within, within vmanage and analytics itself.26:42And then we're also getting application telemetry data. So not only are we making a more informed decision, but again, like with the Microsoft Office 3605, we're getting better visibility into why we made those decisions or the overall health and service of that environment, which is very useful. I think that's one thing that.26:59All of us as network engineers can say is like visibility, that's a good thing, right? And then lastly, you know, we've talked about specifically, those canned, you know, share SaaS applications, whether that be box, et cetera, right? We've talked about using cloud on ramp for Webex and cloud.27:18For Microsoft 3605, but we can also create our custom apps or I mean we do have full n bar capabilities where we can match on over 1400 known applications and do SaaS for that, right? So you have some, so a lot, not some, but a lot of flexibility in creating your own cloud on.27:39On ramp application you want to monitor something else, right? As long as we can match on that or if we don't have that in our standard application list, you can now create your own, right? So I this is maybe private cloud type things or just to you know service over the internet that we want to send those probes to. So as long as we have.27:58Layer three layer four information or fully qualified domain name, right? We can make that decision on that application. So pretty much whatever you wanna do from a general cloud on ramp point of view, right? We're gonna send, we're gonna follow that same standard cloud on ramp SaaS to make an informed decision.28:18Whether it's one of those standard applications or whether it's one that you've essentially created via custom application. So now we'll move into, I'm gonna punt over to Edgar, who is going to go over some demo.28:37Thanks chat. So do let me just share my screen so that you can just confirm if you can see it. You can just confirm that you're watching my screen right now yep yeah see it well.28:55Yeah. Oh perfect perfectly. Okay, so if we go to the I'm gonna go to the configuration steps that are required to enable for SAS, so as you can see, I'm the version twenty.fifteen, that is the version that we are recommended.29:10If you go to this version and you go directly to configuration cloud on ramp for SAS, you will find this dashboard, ok? Very important, if you enable cloud on ramp for SAS in the previous releases of, of SD WAN.29:27So you will see them enabled here, but if you go to the configuration tab, you will see this legend that it says template config. That means that it was imported or it was taken from the previous SD one version, ok? So that means that you will control all of the.29:44Configuration of those the those sites and those devices using this specific configuration, ok? In my scenario, this is the example I have some sites that are configured to use.30:05We can see here that are enabled we have these sites and these devices are enabled and it was confused with the previous one, right? So in this case, if I wanna do any change of this, I will need to go to this configuration specifically. If I want to enable more applications into the into this into the list, I may wanna.30:19Add applications. If I want to add modify the role that the devices have, e.g., the gateway on this guy's the the DIA, I will go directly to these menus and in my case, and if I need to do the custom application list I'm gonna use all of these options, ok? So everything is gonna be in this way.30:37Now if I'm gonna enable.30:57Yeah, new devices or new sites. I'm gonna use configuration groups and policy groups because that is the option that we are handling in this version, ok? So in my case I have one device that is inactive that is actually part of my sites, but it's using configuration groups and it's using policy groups. And you will see how simple it is to enable.30:59Using this new feature, ok? So if I'm in this menu, what I'm gonna do I'm gonna do enable cloud for SAS, I'm gonna select the application that I'm gonna enable in this case.31:19I'm gonna do for office and I'm gonna do for conquer, but it's a couple of options that I have here. If I go next, it's gonna automatically it's gonna read all the policy groups that I already have and it's gonna look for application policy, ok? That means like.31:30In my configuration, my device should have an application policy already enabled assigned I'm gonna I'm gonna go back just to show you, but you can see now that it's reading those options, ok? So in that case I can I I will be able to enable.31:50The because it's gonna be directly applied to that policy. Obviously I can do it directly in the policy if I want it, if I don't wanna do this wizard, but my recommendation is to use the wizard because it's simple, right? If I go back, I'm just gonna exit, I'm gonna go back to configuration I'm gonna go to policy groups. You will find.32:05This main corporation policy group and I have the application priority policy that I was mentioning before, right? This is the policy that is gonna be modified and is gonna be written during during the workflow and it's gonna modify to enable the cloud on ramp for SAS, ok? So that as prerequisite means like do everything needs to have.32:25Your policy group and you already have your application policy configured there maybe with a basic setup, I mean you don't need that maybe or maybe you haven't really some set up.32:45So it's what it's gonna do applic cloud on ramp for SAS is gonna modify this policy during the workflow that I'm I'm showing, ok? So I'm gonna go back to I'm gonna do enablement for us again I'm gonna select my application that I want to enable. I'm gonna select.32:54The application policy that I'm using for that specific device in that policy group. I'm gonna go next and then it's gonna just confirm the the policy and the number of devices that are associated with the same policy.33:14I'm gonna do submit and after that, I will need to deploy deploy this configuration. So I'm gonna go to deploy main corp policy the policy that I have. So I'm gonna select the device, I'm gonna go next. And when I go to this wizard, it's gonna show me all.33:28So the devices that are applicable for canon and I'm gonna select the options that Chad was mentioning some minutes ago, right? So what is the type of site that I'm gonna enable if I'm gonna do the DIA directly from this site or it's gonna use as a gateway just to pass the traffic from other sites, right?33:48So I can discriminate the configuration that I wanted, right? In my case I'm gonna do dia. That is gonna what I'm gonna do. I'm talking about the T logs so also I can control here and I can enable load balancing in any case, e.g., forget ways that it could be applicable, right? So what I'm gonna do is go to this wizard. I'm gonna do good next.34:08And then automatically you will see that it's gonna go to the screen where I'm gonna modify the controllers because the policy's gonna be changed there and it's gonna be pushed. So I'm gonna do deploy.34:28And it's gonna take some, some, some time. Let me go to view deployment progress just to be sure that it is configured and let's just wait for minutes that is apply the policy on the on the controllers, ok? That is what I'm serving on it is finished. I can go back to the to the configuration plan on.34:38For SAS, I'm gonna be gonna take like some seconds. You're gonna see now that my device should it disappeared for inactive. Now it is active here, I can see here and it will update the application that are enabled here because see you remember I selected the Microsoft and I selected the conquer, so it's gonna take.34:58Sometimes just to refresh the the dashboard to appear here and start appearing. What you're gonna get on this dashboard on your enabled, you will see the list of application that are enabled, I mean, you don't click on the application, you will see a filter about which device which sites are using.35:18That, that application, so you will discriminate or you will have a a quick see of what are the applications are running each site. You can see now I have office and I have Webex.35:38If I do office, click on office, I will see just the site that I just enabled because that is the only one that is running for for that specific. If I do look on Webex I will see which other sites are using the Webex or in my case Conquer that is just again my site, right? So that is what I wanna have. Now, once I can.35:53If I wanna modify, e.g., if I want to enable more applications, I wanna add more application on this side specifically I can run the wizard again. I mean, any time that I need it, I can just do enablement for select the applications that I want to add and it's gonna be modified the policy with applications that.36:14I need it. This is one way to do it. And if you are expert and you wanna do it manually, you can also do it manually if you go to again two configuration policy groups and you go to the, to the policy application priority policy, you will see this is my my policy I can edit that and I will see.36:32The rules you know that were created for and I can create my own if I want it, right? Just following the same rule. I recommend to use the workflow again because that is the simple way that we can achieve it.36:52But you can also with do manual things if you need it, right? So something that also is very important is once you have all your applications configured, how I monitor them, right? So I can know what is the current quality of experience that are experimenting on each of the sites. So I'm gonna do click.37:05On the site name directly I'm gonna I'm gonna click and it's gonna reference to the, to the device that is in that location. I'm gonna start watching the metrics that are related to quality of of of experience status and and the score, right? The status of of the metrics and and the score I will have. If I do click here.37:25I will have the graph in the time and I can see what is the current interfaces and what is the the calculated or estimated QA that I'm having for that specific application, right? So that is the way that I can I can monitor. Also I can use the usage, I mean, this is our lab environment. We don't have really traffic going on, but I'll show you do click on the view.37:45You can get some interesting graphs about how many traffic you are passing through and using the different modes that are operating on the sites, right? So it's depend depending on your environment. And now Chad mentioned like we can have advanced telemetry from some application that we are doing partnership, right?38:06So how I can hear that. If I go back to the.38:25If I go to telemetry settings, that is this option in the top, you will see that I can have some options like e.g. enable application telemetry for Webex that was something that we mentioned that it's an advanced telemetry in order that we can have more real statistics and more data coming from, in this case where.38:30Next hop. So if I do click here, it's gonna ask me for the the client ID and second ID that is obviously using the Webex app to obtain that data and use it, right? Or in the case that I'm gonna do Microsoft, it's gonna select Microsoft and the account in Microsoft to do the enable the advanced telemetry and get that data, right?38:50Obviously it's recommended once you enabled your applications and is recommended that the user, right? So.39:10This is the way that you you can figure out, you can see it's very simple, and with this chat, I'm passing to you back to to continue with the presentation. Perfect, thanks and that's great. Actually, I love that seeing the workgroup side of things or well sorry config groups policy groups just definitely helpful as well.39:17Yeah let's jump back into our presentation and we're gonna continue on with cloud on ramp for multi cloud. And one of the things that I mentioned before, cloud on ramp for multi cloud being kind of an umbrella for all of the multi cloud features that we have. So we have multiple capabilities, right? Cause as I mentioned kind of in that 1st what problem are we trying?39:37The soft side.39:57Right, it isn't quite just going to site to cloud anymore or site to or enterprise to cloud. So for infrastructure as a ServiceNow we have capabilities, whether that is, yeah, I definitely still have those workflows, I still need to connect to AWS or Azure or google or Oracle or whatever it might be, right? But I.39:58I also potentially want to do site to site and we can actually, and I'll we'll we'll look at some slides here later that show we I can do site to site connectivity over those cloud providers, right? So there's capabilities there that a lot of people aren't really necessarily considering or they weren't considering definitely in the earlier days of cloud, but now we can.40:18Can use cloud as part of our backbone if we wish, right?40:38And also cloud to cloud, especially for those disparate cloud connectivity options, right? Where, ok, now I need traffic from one cloud to be able to reach another cloud. I can use our our our essentially cloud on ramp capabilities in order to connect those, right? And one way that, that one would do that would be with with our S.40:42CCI, right? Software defined cloud interconnect capabilities and whether or not you're doing this on prem or in many cases, right, with one of our providers like Equinix or Megaport, right, where we're going to use.41:02One of our edge routers is maybe a hub for our different cloud providers where we're going to instantiate those catalysts 8000 Vs to connect to AWS, Azure, google, Oracle, what have you, right? So we use SDCI for multi cloud connectivity. Again, like I said whether you're building it within one of the, these providers, which you.41:14Can do very easily, actually it's it's, it's quite scripted.41:34Or otherwise, right, we're using SD WAN fabric to connect to not just our enterprise sites, but also our disparate cloud providers and even potentially using them as backbones to get from site to site. And all of that would be kind of for not right if we didn't have any visibility into it. And it's one of those.41:38Things that, you know, the cloud obviously being a somewhat nebulous term in general, right? If we just said, ok, I sent it to the cloud, seems to be working, right? I mean that's I guess all well and good if everything always just works.41:58Right, but when you have issues, we need a way to ok try to determine where that issue occurs. And of course within manager, we do collect that data and from the disparate cloud providers try to and try to present that in a single pane of glass so that if we're going to determine or try to determine.42:11How is it provide or how is it performing, right? Is is everything good? I'm not getting any tickets or calls, so I'm gonna assume everything's ok but right that's the moment you start doing that too much is the moment you get a call at 02:00 A.M..42:31Right with some slow burn problem that's been going on for a month. So we do uncover the and gather data for each of the cloud providers that we're using and we present that in SD WAN manager itself. So we get a better single pane of glass visibility into the health of the solution that we've just deployed, which is perfect.42:45Going a little bit further, we're gonna start with cloud on ramp for multi cloud with AWS, and this will be specifically thinking about it as, you know, that infrastructure as a service offering is kind of step one, right? How are we connecting to our vpcs or workloads in AWS? And we do this, I mean.43:07You can still, there's still the manual option of always connecting, but with cloud on ramp for multi cloud, right? We can instantiate.43:25The API calls automatically catalyst 8000 Vs in, in in your environment in AWS, Connect and map your vpcs to whatever VPNs are appropriate for that environment and essentially connect that environment to your fabric all in a couple steps.43:32Apps, right? So, we can quickly turn up connectivity into the cloud without having to manually build all of this, right? The APIs and workflows essentially, we're doing that all for you once you enter the appropriate information.43:52We can also do, as I mentioned, something that we don't always, like I said, this isn't something that's, that's always considered, but we can use that, that backbone as a fabric and whether that's because, based on, you know, if we already have a large cloud presence, I mean there's multiple reasons why we may want to do something like this, right? Whether it's more the.44:06Be able to more quickly turn up a remote site where we can only get internet connectivity and, you know, we can connect more quickly to a local pop and then from that local pop we have a higher speed capabilities you know already in the cloud itself, but we don't have a pop in that location as an example, right? So there are many reasons why we might.44:26Want to do this, but essentially we can extend our fabric across the tunnels that are across the, the cloud itself right in order to form multiple topologies depending on so that we can extend traffic and essentially use the cloud provider's network.44:46As our network in this case, right? It's one of those things being vendor and cloud agnostic for the most part, right? It's whatever circuit we can get and sometimes that circuit is someone else's in this case, right? We also do site to site with cloud wan, in this case, right? We we aren't building a fabric across the WAN, right? But essentially.45:05We are.45:25Instantiating those catalysts eight KVs, creating a BGP session with the their core, the core network itself within our AWS fabric and able to connect to remote vpcs in this case. And we do the same with site to site, right? So we don't have to build the fabric across. We can use right, create our vrfs or.45:26And then essentially use the backbone segment itself to share those those prefixes so that when we are getting to that remote site, right? So our fabric exists out outside of this on on essentially kind of on both sides, right? So, we were just doing, network advertisements within.45:46The AWS fabric itself to use as their, as their background. Both of those are valid deployment models that we can use within AWS. We have a similar capability, obviously, with Azure, right? So we say same as we could with AWS, right? This is gonna be a pretty standard story, right? We can instantiate those catalyst 8000.46:03These automatically via API call into Microsoft Azure, right? And within that peer with vhub.46:23And map the correct vnets to the VPNs that we need in our environment so we can very quickly, you know, same way, instantiate access to the environment and build that access without having to build all of this as you manually and run a bunch of PowerShell scripts et cetera, right? So we can build this all within standard workflows, right? This is.46:30Is.46:50Pretty basic level capabilities that we've had for, I don't know, for for quite some time in this case, right? So this is something that is pretty much table stakes at this point, right? And this, same with AWS and govcloud it's the same with Microsoft Azure and Azure government cloud as well. So capabilities are available in both.46:52This case.47:11And similar, same story and it's gonna look the same across multiple providers, but we can also use Azure for building tunnels across their environment, right? So if we want to use their environment for some of that same capability, right, where we need to extend our reach in places maybe.47:12We don't have it or we just like the the billing model of their backbone better than the billing model of.47:32You know, one of the other standard telecom providers out there in whatever region you're trying to reach, right? So whatever reason maybe, the capability is there and I do have customers doing this these days, right? So we can use the V WAN capabilities and extend our fabric wherever we need to go and kind of use that as our backbone.47:40Excuse me, Google Cloud, again, I'm right, same type of story, right? We're using the terms change a little bit, but the story ends up being the same in this case, right? Which is a good thing, right? How we connect to Google Cloud, this is a little bit newer than AWS and Azure from a capability standpoint, but it's been.48:01In there now with catalyst SD WAN for quite some time multiple releases, right? So we we can again instantiate those catalyst 8000 Vs automatically via workflow, Connect to the the backend Google Cloud registry and routers and.48:19Map the the correct vpcs to essentially to whatever VPN we need and we, you know, we quickly enabled access into that cloud, right? And so with all of these, right, this ends up being another site in our fabric, so whether or not multiple sites, can connect directly with you know.48:36Their standard BFD tunneling in that case or whether or not you then have a topology where this connects to a hub, you can essentially treat it like another site in that regards if you need to, how you create your topology considering this being another hub in that case. With BGP or sorry BGP with.48:56Google cloud, we can also use their backbone as well to create SD WAN tunnels. So same idea, right? Regardless of your provider.49:16You can do basically the same things and whether that's I mean the the standard thing which is I need to connect to the cloud because I have workloads in the cloud and my users need access to those workloads or I want to expand my use of that cloud to enable traffic, you know, for more whether it's just more efficient, whatever the reason might be.49:24The.49:44I can use the cloud now as a backbone provider as well, right? Or just extend the reach. It doesn't have to be your entire backbone. It might just be for that one site that's out there that's, you need access to very quickly for whatever reason. I will take a step now into cloud interconnect or software defined interconnect or cloud.49:45Cloud cloud interconnect essentially, right? So essentially we can do the same thing with our partners Equinix Megaport, right? We can quickly enable services inside one of their pops or one inside of one of their data centers and use that as kind of a software defined cloud interconnect.50:05In this case to connect to multiple cloud services. Equinix Megaport both have high speed connectivity to AWS and Azure so you can very quickly spin up a device whether or not you're doing colo services at these or not, right? You can use them in this case to create that kind of that pop, that centralized.50:23A point of presence for connectivity and I have actually a few customers doing this, right? Where I'm gonna connect to this is essentially another, you can consider your cloud hub in a way, right? And based on that, these will, will form multiple connections to Azure.50:43You know, Google Cloud, AWS et cetera, AWS or Equinix or within their environment, right? So as I mentioned in the previous slides, we can certainly use the the cloud provider's backbone, but in many cases this is more comfortable for someone. They want to detach, I don't want to use AWS, I don't want to use VWAN, I want to.51:00Can use Megaport or Equinix, right? I can use their backbone to turn up a port on either side and essentially quickly extend my reach as necessary.51:19And cloud to cloud is the other option here which we don't talk about. This isn't done as often, but often workloads in one need to reach workloads in the other especially I have some customers that are only cloud 1st, right? They don't have any local data centers, and in situations like that, it's even more common I need to be able to get from here to here, right? Between those two. So.51:30Certainly that is one of our capabilities as well and the this the configuration workflow is actually pretty simple, right? Credential management, create the gateway.51:50Turn up, a connectivity. And with that we will go back to Edgar for demo number two. Thanks Chat. Let me just share it again my screen you just can confirm, you can see my my screen.52:01Yep, I can see it. Okay perfect. Okay, this time I'm we're gonna go from the scratch if we go to the so you will see that as you mentioned, there is a predefined workflow that is gonna give you the steps that you require, ok? Before that, it's very important to know that you need to design in my scenario I'm gonna deploy.52:18Yeah, a couple of K routers to be the edge routers in the cloud. I mean, I need to deploy those visual routers. So in, for that purpose I need to have.52:38Those devices, I mean declared it using a template, ok? So I can go in either way if you are gonna go in a way to use a configuration groups and declare the basic template for those devices is one way to do it. That is probably the recommended one. But if you want to have something very.52:51More straightforward and very simple or as the reference, you can use always the the default default templates that we have in the system, ok? I'm gonna use the default, you can use the configuration group as you wish. In my case I'm gonna use the default I'm going directly to sorry classic templates. I'm gonna go to templates.53:11I'm gonna go for default option. I'm gonna look for the AWS template. I'm gonna found the the the 1st option, right? That is talking about a thousand V devices. So what I'm gonna do, I'm gonna touch a couple of devices there.53:31Now I'm gonna use later because when I do the deployment of the gateway, it's gonna ask me for the view IDs of these devices, ok? So in my case I'm gonna select a couple of of devices that are gonna be here. I'm gonna touch those. I'm gonna go to just configure the basic setup, ok? If I'm gonna edit the device.53:46I'm gonna put the color in this case could be internet or public internet depending off of your setup. You're gonna just put the, the name of the host name that you're gonna assign to the couple of these devices. In my case it's gonna be, this is gonna be the the name. I'm gonna provide the, the IP address.54:06So I'm gonna provide the, the IP address that's gonna be used as system IP. I'm gonna put the site ID that is very important because that means all my AWS connectivity or those devices are gonna be configuring a specific device, ok? So it's very important that you declare that as a, as a new.54:26Device because as Chad mentioned, we are gonna interconnect all all our devices to that device that site. So for us like the AWS site, but we need to declare that, ok? It's very simple, it's as a normal site creation that we have to just create the site.54:47So in my case I'm gonna use the site, this site that is gonna use it, ok? So I'm gonna update this one. I'm gonna go for 2nd. I'm gonna use the same color. I'm gonna use AWS. I'm gonna put the.55:02The IP address, oh then twenty.twenty one. I'm gonna put the site ID again that is gonna be 1011, ok? So I'm gonna update that and I'm gonna probably I'm gonna change this and just put it wrong IP address.55:19So I'm gonna change this one, ok? All right, so I'm gonna do that, I'm gonna go to the normal configure device floor that I'm I haven't that I normally use. I'm gonna confirm, ok? So I'm gonna prepare these two devices that are gonna be assigned in this case to the location that is gonna represent.55:37AWS.55:57Location, ok? So it's done, I'm good. The next step is gonna go go directly to the cloud on prem for multi cloud as you said. 1st I'm gonna associate my account, ok? That is my 1st step. You have all the options here depending on which cloud you're gonna connect in. I'm gonna use the Amazon Web services. That's.55:58The the example I have today.56:18And I in this case I'm gonna just provide an ID, a name, so I'm gonna provide this one. In this case, you just gonna ask for cloud gateway. This is the recommended option that means that we, our SD one solution is gonna do the cloud gateway that is gonna consume.56:20Figure the couple of these routers and it's gonna do the transit vpcs and it's gonna do BGP with internal VPC vpcs and then make connectivity. So that is the recommended one. I'm talking about for you gonna login, you have a couple of options, use EIM role, that means if your your organization is using these kind of roles or using the key that not.56:40Normally the key is the most straightforward because for the roles, you know, the AWS admin needs to provide you the correct group security group in order to give you the correct resources that you need it, right? So in my in our most of the cases is important just to use the key, but it again it's gonna depends off.57:00Of what is your your AWS management, right? If you go for the key, it's gonna ask you for the key and it's gonna ask you for a secret that is gonna be introduced here. Okay, it's a couple of of values that is is gonna be pointing here. You just go to add.57:18It's gonna validate that the connectivity is ok. As you can see now, I have a couple of of accounts that I'm connecting, all both are WS, but I have a couple of accounts that are connecting. The one that I created and another one. Once I finish that, I'm gonna I'm gonna go back to the to the flow and I'm gonna go to to the.57:36Global settings that I'm gonna need to provide, ok? In this case. So I'm gonna edit this one. It's gonna ask me for which type of cloud provider and which type of, what is the account that I'm gonna use? I'm gonna use the one that I created. And here we have different cloud gateways that we can select, ok?57:57The most traditional is gonna be transit gateway using connected, vpcs, ok? This is the most traditional I think show that in the slides. You have also the option to use cloud cloud one that is also.58:16Another option that you can use using infrastructure, that means that it's gonna create a one region in your AWS on and it's gonna provide the all the routing and all the connectivity. And again, it depends off your infrastructure and depends who you are handling your infrastructure in AWS. If you're gonna go straightforward, I'm gonna go.58:32In this way, ok? Or in my way I can also do cloud one because it's allowing my account, ok? So.58:52It's gonna confirm which account you are using, and now it's gonna go go for the region that you're gonna use as a reference just to deploy this, ok? In my case, it's gonna be US east. I mean it's the reference where it's gonna deploy the.59:00The connectivity, the cloud gateway and in the case of software image, I mean you can we already have in the market of AWS some versions are available to deploy there, ok? What is gonna be the recommended eh, I mean depends again of of your environment.59:15Usually we are gonna say that seven.seventeen.fifteen.04 that is the recommended for our edge, I mean in the version twelve.fifteen that we are using, but we can also use another another option if you want it, right? I'm gonna use in this case that one.59:34And then it's gonna select, it's gonna ask you for the instance that you're gonna deploy, ok? Here again depends on your resources. Normally the recommendation will be dependent of how much traffic you're gonna pass through, is gonna be the the what is the correct size that you have. You have several options.59:50I'm gonna go with medium, that is this option but again this depends of your environment and depending of your recommendation, right? So.01:00:10Talking about the subnet pool is gonna be the pool that is gonna be created just to be the transit vpcs, you know? It's gonna be the internal connectivity between your routers and it's gonna be the connectivity that is gonna go with to to do BGP with the, with the host vpcs, right? So we're gonna supply.01:00:19That. What is the recommended here is a a network that is obviously NO.01:00:39Ruleable in your environment. I mean, so you can dedicate and usually the system is gonna provide you this by default. So you can provide that or a specific subnet pool, it's gonna be complete internally the AWS environment. You won't see those IP addresses I mean really coming, right? So just gonna be.01:00:44Something just internal. What you're gonna observe is gonna be through the OMP routes is gonna be the the final, the host vpcs are holding the visual machines. This is just for internal connectivity that is gonna be bring up.01:01:04You're talking about the BGP autonomous system. You can start, you just provide a reference and if it's not if it's in use or is doing some overlapping, it's gonna select another range, ok? In my case I'm gonna use start the the.01:01:18They're recommended and how many tools do you wanna do you wanna bring, right? So we have other options here, e.g., I'm gonna I'm gonna recommend that use the full mesh and use the default settings I mean at least you have any specific requirement, right? But normally we're gonna go for for for the standard.01:01:36Set up. I'm talking about the tunnel. I'm gonna use IPsec, ok? That is what I'm gonna do. So this option again, you don't have mesh so you can select it. So you wanna don't announce the default route, you're gonna disable obviously we need it, right? So that is the reason that I'm gonna go for that. So I'm gonna update.01:01:56This parameter and it's gonna take some time. So once I have here configure and declarate all the details about about what I own in my deployment, I'm gonna go back to the, to the flow.01:02:17And then I'm gonna create the cloud gateway, ok? So I'm gonna do create a gateway. I'm gonna put a num a name for that provide that and I'm gonna select what is the account that I'm gonna use. I'm gonna do do for 2nd option.01:02:32I'm gonna select the region when I'm gonna deploy my instance in this case.01:02:52I'm gonna go through US East one, ok? And then it's gonna read the general parameter that we previously defined, you know, it's gonna take in mind what is the, what is the the size of the visual machine, the deployment of the visual machine et cetera, is gonna be taken in this moment.01:02:57And then it's gonna ask you what is the site where you're gonna deploy this. Remember that I assigned a couple of devices in this site, so it's gonna be taken here, ok? So, once I put it here, it's gonna it's gonna populate some information, ok?01:03:16And it's gonna go and confirm the same parameters that we specify in the previous, right? So only we specify the those ones it will take some time you will find in the in the right section.01:03:33You will find who is the topology that is gonna build. We have the gateway, we have the transit VPC and we have the couple of devices that are gonna be deployed just to make the connectivity, right? So that's what I'm gonna have. So it took sometimes just to read the data.01:03:48And then, you don't need to specify configuration, you need, you need to specify configuration group if you need it, I mean if you are doing configuration groups, you can represent it here or if you're asking a temp you're you're using a template.01:04:04You can select directly from the template here, right? And then you go come to configure, right? So that's what it's gonna go. You're gonna go to the next screen and it's gonna give you it's gonna take some time to deploy, I mean obviously because you need to make the all the API calls and orchestrate the the configuration, right? So how it gonna look once it finished.01:04:20Well, you can go back to configuration on the cloud. You will find this screen that is pretty similar where we saw in the so you can see all the connectivity that I have with the different clouds. So in my case, e.g., AWS, I'm gonna be how many gateways, cloud gateways I have.01:04:40How many connections and and I can go in details, right? So if I go here, I'm gonna be my my akabs deployed, ok? And I can be, see what what is the current connectivity, right? So.01:05:01If I go back, I I will come back. I will see more data, right? And I can see the number of connections and if I wanna use tags it's also another recommended way. That is gonna be to identify the vpcs and how we can interconnect the vpcs that you can also specify during the setup, right? You wanna tag the the.01:05:14The vpcs you can also use that option to to do it, right? So I have this is this dashboard is.01:05:34It's gonna be more control dashboard, ok? It's not just to show you what is going on, what is their current connection, how it's working. You want to monitor the traffic and the real data of multi cloud, you will need to reference directly to the monitor and you will need to go to multi cloud.01:05:43You will have here more details, ok? So here you will find more statistics, how many tunnels, how many tabs you have, and you can do click and do more going more detail and go into each of the devices and find information how much traffic you are passing through, what is the connection, the site connections.01:05:59The gateways that you have associated branches. So you have all the control over here, ok? So it's very important that you reference to moni to monitor multi cloud in this case just to see how it's working, ok? The other panel is just just to configure and just set up the that's the configuration is there.01:06:17But I don't see finished, you will need to come to this one, right? And you will see and again you have some the Megaport notes on our connection, you will also be able to monitor in this section, right? So that is how we are gonna watch it. So with this chat, I'm send it back to you too, to presentation to finalize.01:06:37Perfect. Now, yeah, thanks for sharing that again. That's really great to see especially some of the the aspects that we covered today with both the configuration as well as the monitoring. Let me pull up the presentation again and just to recap everything we've talked about, we reviewed cloud on ramp for SaaS use cases essentially what problem are we.01:06:56Trying to solve and how do we solve it? We've talked about cloud on ramp for SaaS in great detail. Edgar will let us through a demonstration of how it's configured, how we can view cloud on ramp SaaS.01:07:16Metrics et cetera. Then we've jumped into cloud on ramp for multi cloud where we've gone into detail on the infrastructure as a service side of things, the cloud connectivity, so SDCI interconnect capabilities within cloud on ramp, and then Edgar was nice enough to lead us through that great demo of what it looks like to deploy the situation.01:07:28In your environment.01:07:48So key benefits that we want to make sure that we talk about that we enforce, right, why one might want to go and use cloud on form for SaaS. I mean we can go through each of these, which I won't do, but essentially it allows us gives us a strong security posture reduce risk to provide a private backbone or the ability to have global.01:07:51Reach with guaranteed SLAs where we may or may not have that capability with other providers et cetera, which essentially provide solves some of our IT requirements for con you know business continuity.01:08:11Type situations, automation capabilities within SD WAN manager for day zero through day end type orchestration where we can quickly spin up and connect to our cloud workloads or we'll roll out connectivity to multiple sites et cetera, just based on what we're trying to accomplish. And we can do that all within.01:08:22A single pane of glass is the idea as much as possible, right? We can manage that from within catalyst SD WAN manager itself, right? So that's the the benefit of doing it there versus kind of doing it the old fashioned way manually putting it together, which is still possible, but we kind of lose a lot of that visibility that we're able to pull in otherwise within cloud onramp.01:08:42So we have some resources that I always recommend and I always, so you could take a screenshot of this if you like, but a quick Google search will also pull this right up with the 1st if you do a cloud on ramp config guide, definitely recommend this. This is a great one of the better config guides in my mind, right? It really goes into detail for each of the cloud on ramp capabilit.01:09:03Abilities that we talked about today and how how to configure both in via templates, config groups or policy groups or just via CLI if you need to, right? Then also there's the SD WAN design guide I think a lot of people forget about this after they deploy SD WAN, you know, often in the beginning stages when might go out and and read this and I'd definitely recommend.01:09:23Minute when you're getting to understand and know catalyst SD WAN, but this does have recommendations for other features as you start to deploy them and add them to your environment. So definitely.01:09:42You know, this does get updated something regularly, so make sure that you kind of keep that in mind and go out and take a look at it and see what it what it recommends. There's also several Cisco YouTube channels, but there is one specifically that talks about SD WAN and cloud networking. So there's some great data there, some great videos.01:09:53To to walk you through some of this in more detail.01:10:13And then lastly, obviously we want you to continue on your journey with us. So we have that EIS session next week on policy groups and config groups. Highly recommend next Thursday attend that. There's some other great links on here as well that we webinar recordings, so go to the Cisco community page where you registered for the.01:10:17This proves that some great resources there as well. And with that, I want to thank everyone for spending your time with us today. I want to thank Edgar for spending doing some great demo work and Tushar and Eris for being available to answer some of those questions that have been popping up. Alright, thanks all. Have a great rest of your week.01:10:36