Transcript
Welcome everyone to this 3rd session of our expert insights series on Nexus dashboard.00:10Today we will be doing a deep dive on how to configure and manage LAN and SAN fabrics using the controller capability of Nexus dashboard, which was earlier called as NDFC, but now it is an inbuilt capability in Nexus dashboard.00:19I am Pratibha from the customer success team and along with me, we have my peers, Shyam and Ronak, who will also be contributing to this session. Once again, welcome everybody, and without further ado, let us get into the agenda for today.00:36This is what the agenda looks like. We will 1st do a very quick introduction to Nexus Dashboard controller. We hope you had joined the previous sessions where, our team would have introduced you to Nexus dashboard given an overview of unified Nexus dashboard, and also 2nd session that we had on the.00:55Orchestrator and telemetry feature. If not, we will send you the recording links for that so that you will be able to connect the dots.01:16After that we will talk about operations and engineering of the controller feature of Nexus dashboard. We'll talk about automation for LAN, how do you provision SAN fabrics? We'll then go into management visibility and do a little bit of a sneak peek into some of the advanced features before we conclude the session. At any point of time, if.01:23If you have any questions, please use the Q and A panel on the webinar to ask your questions. In case we miss answering any of your questions, we'll get back to you later through an email or if you wanna ask questions live, we also have a 1 h exclusive session for you coming Tuesday.01:43Where we will do a recap of the entire series and you get to unmute yourself and ask your questions live. So please do use that session as well.02:03With that being said, let's quickly look at Nexus Dashboard controller. As you must have all heard it time and again now. Earlier, we had different software images, one for Nexus Dashboard insights, one for orchestrator, and another for NDFC. But now all of these are in.02:14Capabilities within our unified Nexus dashboard firmware, which is four.x. So you could avail the benefits of orchestration, insights or telemetry and the controller from the same page. And for those of you who are new to nexus dashboard or you already have a Nexus dashboard.02:34With controller feature on it, but do not know where to start understanding the various features. There's an option that talks about journey of Nexus dashboard as you can see on the screen. And this is one place where you could see the entire workflow right from what is Nexus dashboard? How do you bring up the cluster?02:54What are the system settings and so on and you could keep learning more about each of these features. So this is one great place for you to get started. With that being said, for those of you who missed joining the Slido earlier, please take out your phone, scan the QR code and join the slido. We do have.03:13Have some questions for you lined up.03:33I'll give it another 30 to 60 s here before we move to the 1st question if you're having any issue.03:36Who's joining the Slido, please call it out on the chat. My team is there to call out any messages that we could help you with.03:55These questions are to understand your current Nexus dashboard infrastructure. How do you use it? Do you know some of the features that exist on the Nexus dashboard controller so that we can customize the flow for this session and also for any of the upcoming sessions and find.04:08Unit based on your needs.04:28Okay then, let's see how many folks have joined. Here is your 1st question coming up. So the 1st question is, did you know that Nexus dashboard can be used to manage NXOS based.04:31Based fabrics. Not ACI, but NXOS based fabrics.04:50Okay, I see we have five, six participants line. Let's keep it coming in. I'm not able to see the attendee count on the webinar. Do you know how many attendees have joined us as of now just to gauge.04:56How many folks have taken the survey?05:16See around close to 40, ok. Yeah. Okay, thanks, thanks. Okay, so out of 40, I see 15, people who have taken the survey. Since this is the 1st question, I'll give it a few more, maybe a minute more here before we move to the next one.05:24Team please join the Slido and take up the questions. We have three questions for you in total, this is just the 1st one. Okay, so most of you do know that Nexus dashboard can be used to manage NXOS fabrics, which is good. And for those.05:42So don't, we hope this session helps you.06:02In understanding that. So with that being said, let us now go to nexus dashboard operations and engineering. So this is divided into three parts, automation, management and visibility. When we say automation, we refer to how your fabrics are provisioned, which is your day zero.06:05And day one task that you do, right? E.g., bringing your underlay, how do you design your fabric? How do you onboard devices to the fabric? How do you provision them and you know how do you deploy the configuration? And then when you think of the overlays where your vrfs, networks, attaching it to interface.06:25So underlay overlay configuration comes in automation or provisioning. And then we will talk about management which is basically your day two operations like image management, how can you take a backup, restore, what about your RMA workflows if you have to replace a device or ensure configuration compliance? How is that?06:45Done. And finally, we look at visibility. How can you track endpoints in real time, look at topology views and a lot more features, endpoint locator, OAM life protect. So there's a lot more features that we have now. So we'll walk through all of that.07:05So the 1st thing that we're gonna talk about is ND automation and as I was giving a brief overview of it. So this is how you provision your fabrics and all your day zero and day one tasks. When we say underlay, based on some existing templates, right? Let's say you're creating a vxlan EV.07:22Which is just one type of fabric in the LAN persona that you can create on Nexus dashboard. But if you're using the VXLAN EVPN fabric, then think of the underlay configuration as, you know, you give the AS number or whether you want the fabric to use ingress or or multicast replication for ingress replication or multicast.07:43For your bump traffic, what are the IP pools you want to configure? What does your routing protocol, which protocol are you using? So all of these information goes in underlay configuration, and most of these information maybe apart from AS number gets pre populated, but of course you have the capability to modify.08:02As it sits your environment. And then when we look at the overlay configuration, as you can see on the screen, you have the VRF configuration, network, how do you attach it to interface? Finally, how do you re recalculate and deploy for those who have worked on Nexus dashboard controller exc exclusively, right?08:22You will extensively, so you will see that for everything that you push to the fabrics, there is an option to recalculate and deploy, which means it will tell you what is the current state of the switches? What is the current configuration that is deployed? What is your intent that you plan to push by making these.08:42Changes and you can get a side by side comparison with highlighted, you know, command line to show you that these are the changes that will get pushed. So many of these is automated instead of thousand lines of code that you have to write, you just specify your intent and Nexus dashboard will convert that into command line commands.09:02Show you what it's going to push and then you push it. So that comes in day one configuration.09:22And this is the fabric builder that we are talking about. So in the land persona of Nexus dashboard controller, there are multiple different types of fabric that you can create. Classic LAN, enhanced LAN, VXLAN EVPN. So for all of these, doing this manually.09:28Imagine if you didn't have Nexus dashboard controller and you were to do it manually, it would be thousands of lines of configuration. But using the fabric builder feature of Nexus dashboard, Cisco brings to you the best practices out of the box. You don't have to reach out to Cisco to say, hey, you know, is this.09:49What's the best practice, should I do this? Should I do that? It all comes built in out of the box and I was just talking to Sean yesterday where he mentioned that for one of the customers who had an existing fabric and they wanted to do a brownfield import to Nexus dashboard, initially customer had a lot of questions.10:08But they could, as they started working with Nexus dashboard, they could see that even for them to complete the migration, to complete that process, Nexus dashboard would keep prompting or giving you the best practices, you know, maybe go fix your HSRP, maybe fix this issue and only when the best practices and, you know, the recommendations are the.10:28Major issues are met or addressed, only then it lets you, you know, onboard that fabric to your Nexus dashboard. So, you don't just have to click a few buttons and hope that things are going well, but Nexus dashboard takes care of every step that best practices that we have gathered through our on field experiences, right?10:49People from tag, from CX, the BU, and customers we have heard from you. So all of those best practices are ingrained out of the box, but of course it is customizable, so you can edit it as per your needs.11:09So now, let's say e.g. taking VXLAN EVPN as a fabric, right? Example in the land persona, so your day zero underlay, this is what you would look like. 1st you create the fabric settings, AS number, replication mode etc. Then you discover the switches, you finally recalculate and deploy as per the best practices.11:24We're not gonna do a demo for this part of day zero, day one provisioning because this was already covered in the 1st session of the series. So in case you want to go and look at a demo of that, please do refer the recording. We'll share the link for that. But with that being said, again, as I was telling you the fabric builder, right?11:46It's not specific to the VXLAN EVPN fabric. You have classic land fabric, you know, where you have Nexus three case, five Ks et cetera, you have your enhanced classic land fabric where you have the three tire architecture composing of Nexus seven case and nine case. And recently we also have this.12:05You know fabric option where you could add your broker switches, which means earlier you had a component called as Nexus dashboard data broker, which was used for span to monitor the traffic to collect information, right? But now, Nexus nine case switches, they, you know.12:25Data broker switches and they can collect, you know, your tap and span information from production fabrics by using this sort of this type of fabric which is called as the data broker fabric, and it can send the information to monitoring tools, your security tools, and so on.12:46So these are different types of fabrics. Of course, we have AI ML fabric, you know, for those of you who want to build an AI workload, you have the normal BGP fabric and between data center and campus we explain EVPN fabric. So data center is where you use your nexus nine case campus is where you use your cat nine case.13:01So that's the difference, so you could even manage your cat nine K VXLAN EVPN fabrics through Nexus dashboard controller.13:21So these are different types of fabric that you have. The component that we just spoke about was Fabric Builder, which means you know step one, you decide what is the template that you want based on that Cisco would give you the best practices, would pre populate most of the fields. You customize it, you deploy. So that.13:29This is the fabric builder, but now Nexus dashboard also comes with a feature called this fabric designer, which means even before you have the hardware or infrastructure in the site, you can still design your fabric. So if you look at that, you know, what is going to be your leaf spine model?13:49Of devices, what is your IP pool going to be going to look like? Your link speed, your cabling. So you can check if all this is supported with each other or not, any interoperability. So you can check all of that by using this feature. So when you create or onboard a fabric on Nexus dashboard controller.14:10There's a small checkbox that you see which says quickly build and cable prior to equipment purchase. So which means even before you place the order.14:29You can build or you can build a design as to what your network is going to look like and you can also download information regarding cabling, assembly, what your configurations would look like. You could download all of these information in terms of report. And finally, there is also an option to, you know, do a bulk.14:39Swap of.14:59Things like serial number, let's say you finally know what is the serial number, like after you've got the device, you've placed the order, you have the correct serial numbers that you're going to use, you know the exact management IP addresses that you're going to use, then you can bulk swap that information to in one go saying that these are my actual serial numbers. These are my actual IP addresses.15:01And then you can transition this to production mode. So when you transition it to production mode, you can NO longer make changes to the designer. But this designer is more like, you know, you run into a lot of questions, right? Is this compatible with this? What optics do I use? What cabling do I use? So this would give you that full picture to make that.15:21Purchase decision and once you get the infrastructure, all you have to do is cabling, but your ND is all set to push the configuration for you.15:41Okay, now let's go to the next question. What type of land fabric do you have? Do you have a VXLAN EVPN fabric? Hopefully your data center one or even if you have catalysts go ahead and let us know. Or do you have the classic land with Nexus 3K5K7 Ks et cetera or do you have the three tired.15:52Architecture with nine case and seven case. Okay, I see some votes coming in. So I see more of a three tire classic land deployment. Okay.16:12See nine of you have cast your votes. I shall give it maybe another 30 s before we move to the next slide. Okay, ten of you have cast your votes.16:30Okay, so most similar percentages I see. Okay, thank you for letting us know what fabrics do you have. Let us now go to.16:47The next part. So we were talking about day zero and day one, right? So you only designed what your fabric looks like. So your fabric settings part is done, but you still haven't pushed any configuration to your switches yet. So we'll talk about what zero touch deployment or power on auto provisioning is, which means you just.17:07Just rack and stack your switches, you cable them, and you let Nexus Dashboard do everything.17:27Now for this, we will have a demo, so let me just play this demo for you. So you go to fabrics, you select the fabric that you've deployed in the previous step, right? Day zero, day one configuration, you go to edit.17:33Fabrics fabric management, and here you can figure the bootstrapping parameters.17:53Andy comes with an inbuilt DHCP server, but if you have something like Infoblox and you want to use that, you are free to use that as well. Or if you want to use the inbuilt DHCP server, you can. You give it the scope, you give it the default gateway address, and this is how you configure the this is how you configure the bootstrap.17:58Lab settings, you save this.18:18Now once this is done, how do you, so this is how you enabled bootstrap right now, but now we will see how do you actually bootstrap that and connect that to a switch. So you go to switch right now, NO switches are in display, you've just racked stacked and cabled, that is all that you've done.18:22So now you will, in the discovery you will do bootstrap. Let me just pause here. So all of these information you are getting because of the cabling that you have done, right? And the gateway is assigned. So from your DHCP, even if you're using the inbill DHCP pool, the gateway gets assigned from that configuration.18:39And the pool range, right? From there temporarily you have an IP assigned so that you could discover the switches, but if you want to change it to a more permanent IP address that you want to be connected to each of these devices configured on each of these devices, you could go ahead and edit that as well. So you add the IP that you.18:59You want to configure on this device, you give it.19:19Again if you want to select an image policies, you mentioned the role of the device. We'll come to the switch role in a bit, but you could add your admin password and this is how, right? So this is the screen that all of us must have seen, right? When you power on the device for the 1st time, you see a lot of logs and then it'll ask you like is.19:23Scope enabled yes or NO?19:43So that's where you select that. And then here, once you have configured this and the admin password, e.g., you have completed the bootstrapping process. And the last one is what if now for bootstrapping rate, let's say I.19:45Have many switches. I don't have a single switch, but I have many switches. So even if I had to follow this process where you click, you add the IP address, you put the admin password, you save, and the configuration gets pushed to the devices. But imagine if you had hundreds of switches, how would you do it? So then there's something called as pre provisioning, where.20:05You can pre populate the information with the serial number and other provisioning information through a CSV file. So we'll see that instead of the bootstrap.20:25Radio button, we click on the pre provision radio button and now we import. So you have a CSV file where you can import your you know serial numbers, IP address configuration, all of that and it gets autoimported. So if you have a lot of devices to configure, then less number of clicks by using the pre provision.20:37Option.20:57Okay. So, if you remember when we were bootstrapping rate, we saw that there was a place where we selected the role of the switch. Now there are multiple roles of the switch, of course leafs, spine, border leaf. These are all very common terminologies that you would have heard, but there are some terminologies that usually confuse.21:08Use, you know, customers who are using Nexus dashboards, so we'll just clarify that in this slide. So if you look at 2nd option here, border gateway, so this is basically any L three device that you use to connect your existing fabric with another data center fabric. Now it can be NXOS based.21:28Fabric, ACI fabric, etc. So it's basically acting as your data center interconnect, right? That is what your standard border gateway is. And if you see on the diagram on the left hand side, you can see that you have a leaf layer spine layer, and then you have the border gateways which are connecting to your, you know, to the fabric on.21:48The other side. So this is your border gateway. Then you have something called as a VPC border gateway.22:08Now this is when you have a traditional LAN or Ethernet network, right? Where you do not have any VXLAN EVPN, but what if you want to connect that as well to a data center which is actually running VXLAN EVPN? If you see the diagram on the right, so we have a legacy site two, and you want to connect this to another.22:14Fabric with VXLAN EVPN, then you use these called as VPC border gateways that understand that can communicate with the border gateways on the other side or if you also want to migrate some workloads from this legacy site to the other side, so for vmotion to work, that is also another use case where you.22:34Might use these VPC border gateways. And the 3rd one is the border gateway spine. Now, imagine on the left side, you have the or you can see on the left side, right? You have the leaf layer, the spine and then the border gateways. But for some reason, probably you don't have the rack space or the requirements to have two separate layer for the border.22:54The Gateway and for the spine, then you can collapse those two layers, the border gateway and the spine layers, so that will be called as the border gateway spine. But one thing to consider is whenever you do a collapse of two layers, any change that you try to make to maybe the spine features or parameters, and you want to do a reload, then you will.23:14Also be reloading the border gateway, so and vice versa, right? If you want to do some changes in terms of the border gateway reload that, you'll not be able to do that separately. So that's just a consideration to make. But if that suits your requirement to collapse those two layers, it is possible to do that. And that is what is called as the border gateway. It's fine.23:34Finally, how do you create a VPC, you know, pair of switches? This is very simple. You just go to the switches in your fabric, go to inventory switches, select the switch, click on actions VPC pairing. Based on the cable connectivity, Nexus dashboard will automatically discover the switch that needs to be paired with. You can.23:55Just select on it and save. There is also this virtual peer link feature, which means now you NO longer need to have a physical cable connectivity for a VPC peer link that can be virtualized through the Nexus dashboard by enabling the slider.24:14So now we were looking at more of the day zero configuration. How do you configure the underlay and then how do you, you know, add switches to the fabric? How do you make it into a VPC pair? So now we come to the overlay, which is how do you create your vrfs networks, how do you attach it to the interface?24:30Then finally deployed. So for this, we have another demo for you. Let's clear this to see how you create and deploy vrfs and networks. So we go to fabric, we have segmentation and security and under that for both option vrfs and networks.24:50So go to VRF, create a VRF. Here you will have to put in, again, VRF name, give it an ID, which is the VLAN that you want to extend through VXLAN, so you could add that information and once the VRF, ok, we'll create one more VRF here.25:08With the same thing, so you create the VRF VRF name, what is the VLAN that you plan to use?25:28And this is how you create the vrfs. So we have two vrfs created now, and now we want to create the corresponding networks. So you go to the networks tab, create your network, give your network name. If you want only layer two, right? Which means you don't want to give You don't want this to be a gateway, you don't.25:34You want to give the gateway IP, then you can have only layer two as well configured. You select the VRF there, in this case since we are having it as layer three, we give the IP address information. Name the VLAN just for your management purpose to understand. So you've created a network over here. This is called VLAN ten.25:54We'll do a similar thing for VLAN 20, but probably let's skip that part. So, ok, we've created another VLAN called VLAN 20 and now we'll see how can we do a multi attached too.26:12Attach this to multiple switches in the environment. So you select the networks, you do something called as multi attached because you're gonna attach it to multiple switches. You select which other switches you want to attach it to. Here you have leaf one and leaf two. So you attach there, you select the interfaces that you want to attach it to.26:27So click on the interface, again you might tell me if I have multiple switches, multiple interfaces, you could always copy paste this information. So you select that.26:47For VLAN ten, which all interfaces do you want to allow and then you click on next and finally deploy. So this is the recalculate and deploy that we spoke about, right? So it happens in multiple phases and finally, it will tell you that if you can see right it's out of sync, so there are 01:21 lines that are about to be pushed.26:57You can click on that to see what is the configuration that is going to be pushed. You also get a side by side comparison. So you just you know specify your intent.27:17Nexus dashboard converts into CLI, you can review it and then finally do a deploy. Okay, so you see it's successful. You can also see it from the overview tab as well where it says configuration.27:28Completed yes, so if you go to the overview tab, you could see there is configuration deployment completed. So this is how you create a VRF and network and do a multi attach. Now, how do you attach a policy to an interface? This is one of the very common issue.27:47Use, right? So here if you see, let's say your server team or your compute team comes to you and says that, hey, I'm not able to ping my gateway. And usually you can't ping your gateway, they just come and tell you, hey, it's a network issue, we are not able to ping our gateway, what do we do about that, right?28:05So here you can see you're not able to ping the gateway, but you from a network standpoint, you know, you know, this computer, this server is connected to this switch and this interface. So how do you check the policy on the interface? Now you click on the leaf, you go to the connectivity tab.28:22And then you know which interface it's connected to. Of course you can discover it from the topology view, but this is just a case where you know the interface, but you want to check the policy. Right now it's configured as trunk. This is not what you want, so you click edit to edit the policy and then you move it to access. You give it the VLAN detail.28:39And then you deploy. So this is mainly to show you how there are interface policies and how you can view and edit policies to troubleshoot issues like simple issues like, you know, hey, I can't reach the gateway, and same thing at every task that you do.29:01Andy will show you what is the current config, pending config, how many lines am I going to push? So you could see all of this and then decide for yourself if you want to deploy the configuration. So now if you go back.29:19And you see this? You will see that your thing is working successfully now. Okay, so before we go to the next part, we have one more question for you.29:38Do you have MDS switches in your SAN environment? So, so far what we have covered is we have covered underlay overlay, how do you configure your fabric settings? Which firstly was underlay and then you.29:58You know, you detect your, you rack and stack your switches, you bootstrap your switches, so your switches have some configuration and then you create the overlay in terms of the vrfs, interfaces, policies, etc. And then you recalculate and deploy. So that's how you get a VXLAN EVPN fabric.30:16Now, whatever we've discussed so far is the LAN persona of Nexus dashboard, which means for those of you who've deployed Nexus dashboard from scratch or even maybe deployed a virtual nexus dashboard, right? You would see an option to select a persona. Is it going to be a LAN or a stand persona? And when you select LAN, that.30:34That's when you would get fabric templates like all that we spoke about, right? Classic LAN, enhanced land, VXLAN EVPN, so all of that come under the LAN persona, but Nexus dashboard also has a SAN persona to manage your SAN. And though I see 77 percentage of you say that you.30:54Who do not have MBS switches in your environment, I would still say hang in there because we have more coming your way. You know, we'll talk through SAN and then we will look at some of the visibility and monitoring of Nexus Dashboard controller.31:14With that being said, I will now welcome Shyam over to you. I will stop sharing my screen and over to you for the plan person of dashboard. Hey, thanks for the thanks for the insightful session. Let me quickly.31:30Start to share my screen. So as you all heard from the Nexus dashboard to manage the data center network, we have three personas, right? Under the land, we have the VXLAN or the classic land one of the questions that came through the Q and A, where.31:48Classic line is what we call as the legacy architecture or the three tier architecture or collapsed two tier architecture where you have access aggregation core or access and core and aggregation.32:08Collapsed. And we also have SAN, which is what I'm gonna quickly introduce. I also understand 77 % of you don't have SAN, but I will be quick but also give you in terms of the insights to know the product much deeper. And we also have Nexus dashboard in the media industry, right?32:20Wherever we have the broadcasting or media verticals, Nexus dashboard along with Nexus switches can be leveraged to provision the networking as well as facilitating the, those features as well.32:40With that said, a quick jump into sand management, and just to clarify, which already did, it could be either land fabric or the SAN fabric. It cannot be both and hence when you deploy, you'll have to choose the SAN fabric and that's when you start to manage the sand fabrics. Now, as.32:55As you all know, there are two market leaders in the SAN. One is Brocade, part of the Broadcom and then we have Cisco in terms of the MDS class switches.33:15Now, what we are trying to focus is where how can we manage the MDS switches? Just like a, just like you have data center switches such as the data center Ethernet switches such as Nexus nine thousands, seven thousands, 5003 thousands, as well as Cisco ACI. We have MDS.33:26Switches for the sand part.33:46Now, IBM also resells MDS, so some of you might see certain IBM switches. Those are under the hood MDS, be it from Cisco or from the IBM, we can manage both of them with this persona. So the these are some of the quick snapshot of what are the switches that we support. Of course on the left.33:48We have the primary MDS switches which are dedicated sand switches.34:08And we also have Nexus switches that acts as a unified switch. As you all know and aware that Nexus 5000 and few other series switches can handle FC and Ethernet with the technology called FCOE, fiber channel over ethernet. Now, these switches especially Nexus five thousands, can.34:13Get FCOE traffic. It has the ability to split the traffic into FC and Ethernet and eventually provide the FC traffic to the SAN array and also provide the Ethernet to the core or different layers of networking. We support that as well and then we also support few other.34:33Devices, namely.34:54Cisco UCS from the compute standpoint, because we look at SAN as a way that interconnects the compute to the storage, right? SAN is in between. It's the sandwiched one. On the left you know what we call a Cisco UCS, for some of you, if you're not aware, UCS is unified computing systems. It's cisco.34:56Course computing solution where we can.35:16Host any application you can think of, including the AI workloads. And we also support integrating Nexus dashboard with some of the leading storage vendors, namely Dell MC and, you know, other vendors as well. So all of that we support in the sense of giving you the visibility end to end, right? That.35:18That's the support standpoint. Now let's also look at some of the key fabric capabilities.35:38Just as what Pratibha described for the land management, for your SAN management, we could do pretty much all of that. We give you the ability to discover fabrics. We give you the ability to visualize them a lot better with lot of metrics that MDS and other switches collect. We're also able to perform.35:45From.36:05Certain tasks that can observe the statistics and give you the visibility which can be over emails or when you log in into the dashboard itself. And we also provide configuration management. Just like how in the land world, you have vrfs, you have VLANs, we have interfaces, similar.36:06We could configure the interfaces, and in the sand world is primarily a zone sets and zones. So we do support a lot of different ways to configure zones for, the SAN.36:26Network. And part of that configuration is also very important is zoning. As you all know that we have a single active zone set and then we have multiple zones within. And just like in the land world, you have VLAN, we have something called vsan in the SAN world, which is created by Cisco. Cisco is the only vendor who has it to.36:40To Segregate as well as reduce the broadcast issues, so we could manage all of that in this particular persona. And finally telemetry and analytics.36:59You might have attended another session last week which focused on how nexus dashboard gives you visibility into the data plane and gets you a lot of metrics that you could help to reduce the time to troubleshoot. Similarly for SAN, we have something called a SAN analytics, which is on the switch level and on the ND level.37:10Well we call this as SAN insights. We will also look into that how we give you certain you know weapons that can help you troubleshoot your network much more effectively, and a lot of times what we also provide you is the ability to prevent an issue before it occurs. So we'll discuss that.37:29When we get to that slide.37:49Alright, so I'll quickly breeze through. I don't wanna take a lot of time on the slides, but of course part of the fabric management, we are able to re visualize the network. We're able to look at every port, every ISL or inter switch link fabrics and interfaces and give you the statistics. We also give.37:51Give you the ability to manage your vsans, create multiple vsans, ability to create edit zones zone sets, and also we give you the overall inventory to visualize as well as ability to periodically send them over email to get notified as well.38:11In terms of monitoring and alerting, we do have a a ability to look at these metrics via custom dashboards. We're able to customize the dashboards, the dashlets. We're able to also give you the ability to customize alarms. We have the ability to create alarms. We have four.38:29Broad categories and you could set what will be the threshold that you should be notified of.38:48And hence it's absolutely customizable for your environment. And we also have integrations with Syslog and SNMP to send these information to some of your monitoring tools that are centralized. And we can also provide you and look at what is the baseline and if you're deviating from that and create an anomaly.38:54If you do that.39:14Let's also look at reporting and compliance. You know, we'll, we'll look at this in terms of nexus dashboard overall, but we also have the ability to build reports for your switches. It can be port level, interface level, switch level. Very important thing is also if you're managing your MDS switches via CLI, you.39:16Could see how much of a command line, you know, keystrokes that you leverage and also your backup strategy and also how can you push something for.39:351050 or hundred switches at one time. So all of that becomes easier with Nexus dashboard for SAN management. We also track the audit. We look at all the change management aspects from the ND and also we look at compliances in terms of are we compliant? Are there any thing that were that we're deviating?39:46In terms of the switch hardware, software, and all of that is also provided to you all from a single pane of glass.40:06Let's look, look also into the integration capabilities. As I already said, Cisco UCS is a computing platform. We also have on top of that a SaaS, and optionally an on prem solution called Cisco Intersight that helps us.40:14Aggregate all the compute domains into a single pane and provide a lot more visibility and functionality. We can also integrate into inter site domains when your SAN compute attached to SAN is part of that. We could do that. We could integrate into vCenter for getting more visibility into the.40:31The the.40:51The storage aspects. And as I already mentioned all the meeting, all the leading vendors of storage in terms of your we could integrate and we are able to provide you a lot of best practices and certain features that help you look at the overall end to end connectivity and also mitigate some issues which.40:52Which I'll be explaining doing a demo.41:12Alright, those are just few slides that gets you an idea what can we do for the sandworld. So let me quickly also show you the demonstration of some of the things that we talked about before we get into the land segment, right?41:15Alright, so what I've logged into is the Nexus dashboard, which is of a person of SAN. Now, as soon as you log in, we're able to look at the overview, right? Now here is where we are able to look at how many switches do I have and what is the fabric, looking like in terms of health? I see one of.41:33My fabric or one of my switches is critical, one is healthy, so obviously my focus will be on the critical one. There's also the NPV links. I assume people who are already managing the SAN fabric would understand what is NPV, your nodeport virtualizer, what is NPIV, nodeport ID virtualizer, and all of that. So we're.41:53Able to look at all of those aspects, and we're able to we're able to show you in a single UI in terms of the switches, the switch statuses, also the release that we are running. We are running a pretty old but you you could think about where we're able to if you have tens and hundreds of switches, the ability to look at.42:13All of those software versions and also the models will be extremely powerful. And as you further go down, we will look at details in terms of the modules, the interfaces, and also some of the metrics that we are collecting, right? Now you can see the port usage, what are the kind of.42:33Ports that we are leveraging. These are pretty old lab hardware and hence we're using eight gig, but most of our customers are using 32 or 64 gig, and part of that product line, 32 and 64 gig is where we can give you the telemetry, the SAN insights or SAN analytics, so hence, just a quick call out of that.42:53We also provide you congestion details if there are, in my lab environment, we don't have a congestion, but those of you from the sand world know what is something called a slow drain.43:12You know, the sand fabric works very differently in terms of the Ethernet fabric because here it's the you know we don't have a lot of buffer credits and it's not based on the TCP handshakes. So, when an initiator which is the server and the storage, which is the target starts to handshake.43:25Sometimes the rates don't match and then one part will start to get congested. Now that is commonly referred to as slow drain and we'll also discuss few other features that ND nexus dashboard provides to mitigate those. As you all be aware that.43:45ISL or inter switch links because sometimes we have the director level switches and the access switches and mdss that connects to each other for creating the array and also distribution aggregation. So all of that will will be seen here and also you can see top congested isls where.44:05There this is primarily going to some of the compute end as well.44:25If provided there are issues in terms of the endpoint side or you know fiber channel over over IP, and we can also look at the CPU, the temperature, the power related attributes all from a single pane of glass.44:30Now we can also look at the topology, you can see that there are two fabrics added and each of it is connected to respective compute as well. I've not added any of the, you know, the array or the storage A in it, but if you do, then you get lot more information. As you double click on it, of course you get a lot of information in terms of the health.44:45Status and how it's connected and.45:06And eventually my lab environment is a smaller one, but if you have a larger one with lot of interconnections, then you can see all of that in a single UI. Let me also take you to some of the actions that we could do on the fabrics just as what explained on the Ethernet of the land side. Now we also have.45:09Breaks and few other informations. I'll quickly breeze through them.45:29To give you an idea of how we are, we have a parity of the way we exercise extremely good vision as well as the execution for land. We have the same for the SAN as well. Now, if you go to fabrics, this is where you add your fabrics. You can add a fabric, it can, you can bring in your.45:32MDS switches. Now when you do that, we can also link the compute part of it, which is right now we support only Cisco compute solution. It can be either via intercite or it can be managed from the UCS manager mode. And if and also we have the ability to attach non.45:52Cisco which is the IBM switches, so that is a possibility as well. Now, all you provide is username and password and then it goes.46:12Looks at the overall topology using the cabling just as was what Pratibha explained for the land part and gets you all the devices that you could manage from this fabric. Now once you have created the fabric, you can go inside the fabric. Once you click on it on the right side you see the health status, which we already saw on the dashboard.46:20Now once you click on the pop out option, here is where we can look at the fabrics information much more detailed, right? Now, a few other things I would like to talk about is in terms of backup configurations where we could, you know, create a backup which can be either right now or also we can.46:40Create the abilities to create a schedule for backups so that we back up all of the configurations accordingly. And in terms of the interfaces, we can see all of that information right from here and you can also take actions accordingly. E.g., if C ten, if you would like to do or.47:00Add into a membership or vsan or you would like to shut down or NO shut, you could do all of that right from here and hence you're trying to go away from a CLI based method to the GUI based. And also we have vsans, right? Now you can see there are two vsans available and once you go to the respective vsan, you will also see.47:20A lot of information of which switch does it belong to and what's the vsan membership, and what's the domain ID as well. These are the information that we can view here, but as I explained on the slide, we could also configure the zones and zone set and vsans all from a single UI.47:40Now if you need to create a new vsan, you can create that from here, right? That's the equivalent of creating new VLANs and then eventually you can do all of that from a single page. Now, you know, we're also looking at monitoring of the configuration so we can look at the the overall health.47:58With the overall inventory status and also what's the baseline configuration. E.g., all the command line information that goes into the MDS is all captured here and you can also see what we have been what we have pushed so far, right? Any deviation from this from the outer band will be notified as well. Now there are a few other things I'll probably discuss.48:18This in the the next segment in terms of visibility part, but I would like to also show you a few other things that we can do with the fabric onboarding and management.48:38Now, a lot of times we also know that when you have, let's say ten switches, 50 switches, nexus dashboard can manage up to one 60 sand switches. Now, if you like to upgrade it, how, how about I give you an option of where the 1st option is you use a CLI method where you upload into each each of the individual switches.48:50And then you'll have to upgrade it based on the maintenance window you have and the strategy you have. Or 2nd option is use a GUI based mechanism Nexus dashboard, where you can upload the image and you could push this to multiple switches based on your strategy. It can be, we all know that in the sandworld we have multiple parts.49:10To reach initiator and target the best practice is to have at least minimum four parts, right? From one HPA, so meaning a host will have around eight paths to reach. So accordingly you can have reduced throughput but NO downtime upgrade accordingly, where we could add the images of what MDS image that you're trying to upload.49:30And then eventually you can push them to all the switches and have them run concurrently or sequentially the way you like it.49:50Right? So that's the other aspect. Of course, the process is simple. You upload the image here, we create a policy of how we would like to run it. And eventually we select the respective switches, and we will select, you know, select the configurations and then push the way we would like to, right? That's the other.49:58Other aspects from the.50:18This the the fabric management itself. And I think I wanna also talk about a few other aspects, but we'll talk about that under the visibility section. So with that said, I think it's a good introduction in terms of what we could do for the SAN fabrics, and also.50:22So, one important thing, right, before we conclude this segment of management, let me also show you the zones that you could create from the zoning options on the manage tab. Here is where it's gonna you have two options, either IVR zoning or the regular zoning. Now, based on that, we can.50:41When you select the right vsan, it's gonna fetch the existing zone sets, right? Now, once, you know, it loads up, you can see the zone set that I've already created, but you could also create a new zone set. But as you all know in the SAN world we can have only one active zone set. Now, once you go to the respective zone set, you can see the.51:01Information right here in terms of edit zoning. Once you do that under the zone set, we could start editing zones if they're existing. Right now I don't have any zones, but you can also create a zone, right? Now, as you all know, a zone consists of initiator and target with the WWPN of both. Now we can do all of that. You can create the.51:21Numbers and add them into their respective zone and activate them and hence, which is gonna create the path for the initiator or the server to talk to the target which is the.51:41Storage area, right? So you could do all of that right from here. Now, there's also one more important topic to talk about. Now we also have the ability to zone migrate, right? Meaning if you have a broken switch from Broadcom, now you might be having a refresh and you want to refresh.51:53Into Cisco MDS. So we also give you the ability where you can provide the brocade switch IP address. It's gonna get all the zone information from the brocade, and it's gonna help you.52:12Translate that and then eventually migrate that into Cisco MDS. So your refresh into a modern fabric with a lot of capabilities becomes much more easier where you have automated the migration as well. So that's another thing to call out in terms of the manageability as well as the.52:23The day zero day one activities.52:42With that said, I hope this was a good intro into the sand world and what Indy can do, and I'm gonna give it back to Pratibha and then I'll come back later for the visibility part. Over to you Pratibha. Thank you so much, Sham. That was a great introduction to how we can manage our SAN fabrics through Nexus dashboard.52:45Let me share my screen.53:06To jump into the next segment, I hope you're able to see my screen. Thank you.53:09Okay, so in the introduction, we were talking about how Nexus dashboard operations and engineering is divided into three pieces. The 1st one was automation. We spoke about provisioning, right? How do you create your fabrics, add your switches, deploy configuration, and so on.53:272nd and 3rd is what we will be looking at now. Second is your ND management and your 3rd one is ND visibility and monitoring. Now, when we talk about ND management, we are referring to configuration compliance, image management, how do you upgrade, install patches?53:46How do you back up and restore your entire fabric or maybe restore a part of your fabric, and hardware replacement or RMA rate? So this is something where.54:06A lot of our customers, right? A few years ago to just replace a device would mean a lot of manual steps, but we will see through our demo how now through Nexus dashboard, replacing a faulty leaf switch or a spine switch is very easy almost similar to the power on auto provisioning feature.54:18That we spoke about earlier. We'd also talk about RBAC so of course you can have authentication as well, but nexus dashboard comes with some built in rules that you can assign to user for our back and Nexus Dashboard also serves as.54:38As a platform for managing some of your non Nexus platforms. Like e.g. Catalyst Center is one that we spoke about and a few other platforms as well, which we will look at it. How Nexus dashboard can be your single point of management for all of these.54:58So the 1st one that we will look at is configuration compliance. So what Nexus dashboard, like when you use a controller for configuration, right? So here's the thumb rule. All the configuration comes from your controller because your controller needs to be aware of what is there, what is configured, only then.55:18And you know anything that the controller, you know.55:38Has to suggest that sort of best practices anomalies etc. All that is, you know, controller is the single point of control in terms of configuring your devices. But of course, many times we've seen where, you know, maybe the baseline configuration gets pushed from the controller, but because somebody's comp.55:41With CLI or let's say, you know, there was aP1 situation and somebody who loves the CLI wanted to fix the issue quickly and they did that. So then Nexus dashboard NO longer is the one that pushed configuration to your devices, right? And.56:02Through Nexus dashboard, we often reuse policies that can be used by many switches in the fabric. But what happens when such a drift happens? So Nexus Dashboard continuously monitors your configuration on the switches and validates it with the intent. Intent is one.56:21Whatever we click and configure on the Nexus dashboard, right? That is our intent. So between our intent and what is actually there on the devices, Nexus dashboard constantly monitors and in case there is a drift, nexus dashboard sees that you know hey something else is configured on the device, but the.56:41Intent as per my database is something else, then it'll showcase this as a drift.57:01Now this drift can be mitigated through a sync, so you can do a re sync, and when you do that, a back push is also possible, which means from your switches to the controller, but it's usually recommended not to do this back push, but to you know make sure that you can figure, do your confi.57:06Migration from the controller so that you know your intent is right intact and you don't have to do any back push from the switch to the controller.57:26And this resync happens at, you know, a timing interval of around 24 h, but it is also customizable, but, highly recommended that you check if you really want to customize it as per your needs, but currently the timer is configured to 24 h.57:36So this is one in terms of configuration complaints. We will see the next one, which is a streamlined image and patch management. So I'm sure you would have all heard of image upgrades as your software upgrade, right? The main.57:54Piece of what is running on your device for it to function. But you may also have also heard of smus, which is maintenance, right? Some patches that you have maybe a bug that is hitting and then Cisco releases a patch that you can do to just fix the bug. We also have something called as EPLD upgrades.58:11For those of you who are wondering what EPLD is, it stands for electronic programmable logic device. But leaving that aside, what does it actually do? It is a low level firmware for your hardware. What does that mean?58:31So, you know, usually like back in those days, sometimes your hardware is capable of achieving a certain functionality, but the software is not in part yet. So, even though your hardware can support a certain feature, your software is still not there. And then Cisco releases a software.58:47Where now if you install that software, let's say on your line card, e.g., then your line card will now have that added functionality. But for this, you would have to do an additional reload for an EPLD upgrade. But through Nexus dashboard, you have a workflow where all of these upgrades can happen together without having the.59:07Need to do multiple reloads. So let's look at what the flow looks like. Firstly from the Cisco software download page, you download that image that you need for your switch upgrade. Once that is downloaded, you upload that image to your Nexus dashboard.59:27Once it is uploaded, you create something called as an image policy. Now what is this image policy? So you can choose what is the platform, what is the release you want to go to, and how do you attach and also let's say you want to group the switches, right? Usually we recommend that you do a.59:46Or an even grouping. So let's say you have, you know, a fabric of.01:00:06Two spines and let's say four leaf switches, then you put your odd spine and your odd leaf into one upgrade group. You put the even spine and the even leaf in the other upgrade group, and then you do one group upgrade at a time so that you have traffic redundancy from a network standpoint. And especially if you have vpcs consider.01:00:11Good, then one device that is part of the VPC gets upgraded in one batch and the other one gets upgraded in the other batch. So those sort of policies you can create.01:00:32And then you will attach the image policy to the devices that you want to upgrade, then staging the images, which means which is nothing but adding the image to the device to the switch itself. So Nexus dashboard will push the image to the switch. And we've seen this multiple times in the past where, you know, you have a window.01:00:42You have a maintenance window of a stipulated time and then you're taking a lot of time figuring out, hey, how do I copy this image? Okay, this is not working. What else do I do? So this is a way where you can stage the image well beforehand because of late images are like what certain you know GBs, right? So you could stage it beforehand so that in the maintenance.01:01:02Window you save a lot of time.01:01:22Once you do it, you can validate if the upgrade is going to be a disruptive or a non disruptive one. Nexus dashboard will run a certain set of validations and give you feedback. But also it depends on what kind of upgrade you are doing, right? So we have ISSU, enhanced ISSU, and so on.01:01:25So based on the kind of upgrade also, the validation parameters would vary. And then finally you go ahead and upgrade the switches. So this is what the workflow looks like from Nexus dashboard. We'll now look at a few demos to see this in action.01:01:44So this is how you create the policies. So you go to manage, go to your fabric software, and you can see right now we are in the NXOS, we also have CIOs XE there, we'll come to that later, but you have NXOS and ACI fabrics. Right now we are looking at the NXOS, and we spoke about the EPLD image.01:02:00Any other maintenance images that are there, you could see that. In this fabric you can also see some cat nine case as well. You have cat nine case nexus nine case or Nexus three case. On the images, again, you can see the images, right? For cat nine case nexus nine case, 8000 V So for all of that, you see the image.01:02:21Which is over here. Again, this is another indicator that Nexus dashboard can also manage non NXOS devices.01:02:40So now these are the image policies that we have. Let us create a new policy. So we will give it a policy name. We will select the platform. Right now we're gonna upgrade Nexus nine K or three K select the release that you would like to go to. And like I was saying, if you have packages, right? Like let's say some patches that you want to fix.01:02:48For EPLD that you want to do, EPLD upgrade that you would like to do. All of that can be done from a single go. You wouldn't have to do multiple reloads for this. So you select whichever packages that you want to add as part of your upgrade.01:03:08Then you save that. This is how you create the policy. Now after creating the policy, we will have to assign the policy as well, right? So for that, we go to the same place, you go to the fabric that you are planning to upgrade and then you see a prepare option on the right side. So then you could go.01:03:22Go to prepare, you have either attach policy or detach policy, now we are going to attach the policy. This is the image that we are going to use and then we stage now staging is that option where you can.01:03:42You know, proactively push the config push the software image to your switches, and if I just go back a minute, let me just show you this as well. So you can see once it's added, you can see that it's out of sync. So out of sync is because you pushed a policy right?01:03:54Now right, so every time a Nexus dashboard controller, whenever you create a configuration change or policy, we will have to recalculate and deploy, right? So until then it would always show up as out of sync because the policy is pushed. In this case because it's an image policy, it still shows you that there was a policy that was created and attached. Now you can check the.01:04:14Scales of the plan. You can update the plan as well. How many switches are to be updated? It says plan update is yet to start now assigned groups. So you have either role based or or even based, right? Right now I think it's role based you can see all the spines in one group. So this is role based, but you could also do.01:04:34Odd even based and if you use one of the groupings that Nexus dashboard provides, the Nexus dashboard will never put two switches in the same VPC pair in the same group. If you use the, so this is the odd even grouping. Again, if you look at the odd even grouping as well, you see the oddle.01:04:54Ford the border gateway and not spine and then you see the even grouping in the other group. So this way your network redundancy is retained while you are doing the upgrade.01:05:14And, ok, so now you can either stage it or you can go ahead and install the update. And I was saying if you use the, you know, the grouping that is auto provided by Nexus dashboard, then it will not add any two devices of the same VPC pair in the same group. But if you are manually create a creating.01:05:23The grouping based on your requirement, right? You know that, you know, this is what actually provides redundancy, this is still fine, I can club it together. If you're doing that sort of a thing, nexus dashboard will still warn you that, hey, you know, these two devices belong to the same VPC pair and they are in the same upgrade group. So you might see a potential.01:05:43Outage. So nexus dashboard will still warn you, but it is still possible if you think it suits your requirement.01:06:03Oh excuse me. Now we look at some of the upgrade models. Again, we're not going to go into the depths of it. There is a blog, if you see on the bottom of the screen, there's the link so you could read through that blog for details, but in this slide, what we'll essentially look at is you have based on you see.01:06:10The super spine layer spine layer leaf, and then you see the endpoints. So based on the model of the platform, there are different upgrade options. You see GIR, you see SMU, your patches, ISSU, right? So GIR is basically graceful insertion and removal. So what that means is the device will basically advertise.01:06:30Towards neighbors that NO longer use me as the path. I'm NO longer a valid path through route poisoning, so then what would it mean? You could do anything on this device side because you've advertised that, hey, don't don't consider me in the path and you've poisoned the routes that are being advertised from this device. Now you can do literally anything. Disconnect the cable, make configuration change.01:06:50Change it is almost like that device is not even there in your fabric. Then you have ISSU, which is in service software upgrade where you don't have any data plane downtime during the upgrade process. We also have something called as EISSU, which is enhanced ISSU just for your information. You can read the blog, but.01:07:10There even the control plane downtime is limited to around 3 s, so it's very less so that's what it's called as enhanced ISSU. For those of you who are planning your upgrades of your nexus dashboard the fabrics that are managed by your Nexus dashboard controller, I would strongly recommend read the blog, understand the details and choose.01:07:30The upgrade, you know method that works for you.01:07:50Now we come to what is called as RMA replacement. So this is one of the features that a lot of our customers are excited about. You know, when you, when you know that your device is faulty, how can you easily replace it with the Nexus dashboard? We'll watch a demo for this.01:07:54So as you can see here, leaf two is showing as a configuration status not applicable. Discovery status is unreachable, which means leaf two has failed for some reason and now we want to replace leaf two. You can see the serial number of leaf two ends with GW. Now we are gonna replace it. So you go to actions more and provision RMA. Now.01:08:14Here you've already got the RMA device, you have cabled it. So cabling is present. So you put your admin password and this screen looks exactly like your power on auto provisioning screen, right? Where you have the IP address and because you are doing this provision RMA, whatever persona was there in your faulty device, you would have removed the faulty device and plugged the.01:08:34The new device by now. So whatever persona was there in the faulty device, it gets applied to the new device and all you had to do is a few clicks. Now you will.01:08:54Slowly see this is the CLI, right? Anytime you power on the device for the 1st time you see it power on auto provisioning happens and then it configures the devices with the persona that was there of the previous defaulty device. So you can see discovery statuses.01:09:04Discovering sync is pending and soon enough you will see the device come into in sync status and the serial number has changed automatically. Earlier it ended with a GW if you notice, so serial number has automatically changed. So replacing your hardware through an RMA is very simple.01:09:20Now we'll come to role based access. Again, as I was saying authentication is still fine, but Nexus dashboard comes with some RBAC roles. Access administrator is someone who can make changes to maybe L two interfaces, create some L two port channels and so on, but they cannot make any L three changes to the device. Device upgrade.01:09:37The administrator is somebody who can create the image policy that we just spoke about, right? Grouping which hardware, which release and store or so on. Network stagger is like a newbie in your team. You just want this person to copy the image to the device, but not do anything else. That is the network stager for you.01:09:57Network operator is just a fancy way of saying that read only access and network administrator is someone who has all the privileges, but usually NO, we recommend the customers not to use the nexus, I mean the network administrator role unless absolutely necessary because you have full privileges.01:10:14Next, we will look at the backup and restore workflow. So how do you create a backup of your fabric? There are multiple ways you can do it. So you go to fabric, actions, edit fabric settings. If you go to fabric management, you can see from the configuration backup, you can only take an hourly backup, which means every hour it will get a back.01:10:33Or if you want to schedule it for a specified time, then you can say that, you know, at 12 h every day, please schedule a backup. So this is also possible. If you go to the overview, you should be able to see, ok, you can also do an ad hoc backup. So this is another way that which through which you do a backup.01:10:53You go to actions and if you want to take a backup of the fabric now, so you can do a one time ad hoc backup as well. Now that we've taken a backup, how do we use this to restore it?01:11:13So let's look at an example where you've configured something incorrectly and because of that there's a issue at the network level. E.g., on this slide, you can see you're connecting a new computer hypervisor, let's say a server to your.01:11:24Fabric and this is what you've configured, but there is an error in this configuration. So now you want to restore a previous version of a backup that you've taken. So you go to actions maintenance, restore fabric, even though the term says restore fabric, it's not gonna push updates to your entire fabric. So you.01:11:39Select one of the backups that you have gathered. You can see if you see the delta config in the last column, only leaf one has twelve lines of delta change in configuration. So you are not going to be pushing configuration to all the switches in the fabric, only twelve lines and then you restore hinting. It'll do some validation and then it will.01:11:59To restore it to leaf one.01:12:19And again, like anything you do on Nexus dashboard through the controller, you will know what is out of sync, you will be able to preview it and deploy just that. In the interest of time, I will pause this recording, but this is what that is. You just deploy it and then you're, you'll be able to see that the configuration is fixed on the CLI.01:12:22Again for NXOS for non Nexus platforms, as we were saying catalyst, I mean catalyst nine case, right? If you have a VXLAN EVPN fabric, then you could manage that as well through Nexus dashboard. This is not for customers who are using the is different, but if you do not want to use cat C and you have a catalyst.01:12:41Then you could use Nexus dashboard to monitor that. If you have iOS XE on the edge XR, so some amount of monitoring management capabilities is available or even for 3rd party devices to monitor health of the device, you know, uptime, interface state status and so on.01:13:03Now we look at some of the visibility and monitoring capabilities. The 1st one that we are going to look at is something called as operations administration and maintenance. So this is a feature that comes as part of connectivity analysis. If you, if you go to analysis hub on your nexus dashboard and connectivity analysis for your NXOS based fabrics, there are a.01:13:19Lot of tools, trace path, ping, et cetera, that will help you show, you know, what is the forward path, what is the reverse path and so on. Because you have a VXLAN tunnel, you will know this is the destination I'm the source, but what exactly is the path that is taking in the tunnel to be able to see that you have this suite of tools.01:13:39For NXOS based fabrics.01:13:59Then you have the topology views. So, as you can see, right, for any fabric on Nexus dashboard, you get a detailed topology view. You can start from the topology, double click every layer. You can double click the switch, go there, double click on the interfaces, check what is applied on the interfaces, and even if you have.01:14:01Virtualized workflows, right? Or workloads. If you have Vmiss and you have a vCenter integration on Nexus dashboard, then you will have more visibility into that as well. So vCenter, for those of you who have vCenter, you can integrate vCenter to your Nexus dashboard. So that way Nexus dashboard, you will have entire visibility of your, your home.01:14:21Those your distributed virtual switches, what is the interface configuration, what VLAN is being tagged on those interfaces, you will get all that visibility. Usually we have this scenario right where the compute team says, hey, there's a network or the server team, right? They say that some I'm not able to ping the gateway, it's a network issue.01:14:41Very common thing that they say, right? But through this vCenter integration, you will be able to see what is the VLAN that is tagged, maybe there is an incorrect VLAN tagging and call it out. So if all, if any of you have vCenter and you have a Nexus dashboard, I think the 1st thing that you should do is go to the integrations tab and just add your vCenter.01:15:01Details there so that you would have enhanced visibility of your virtual environment.01:15:22Then you have endpoint locator. This is also for NXOS based fabrics. So you can have real time view of, you know, where is the endpoint, is the endpoint moving from one, let's say leaf switch to another because usually when there are multiple V motions done at a very small period of time, you will see.01:15:27Application performance issues. So you could see where is the endpoint connected connected and details of the endpoint information.01:15:46This is what it would look like. Again, this is under analyzing the analysis hub you would see endpoint visibility and you could see more details on your endpoints. Again, as we were talking vCenter can be integrated. If you have Kubernetes pods, that integration is possible. We now have slum integration, if you have AIML.01:15:54The workload, the so to monitor your GPU utilization, if your AI workload is, you know, having some performance issue, is it because of GPU, something on the optics, something on the network? So all of that is possible to be monitored from one place, which is your nexus dashboard and we send.01:16:14Integration, so this is where you have to go and your integrate your vCenter.01:16:34Finally, we come to two last things before I pass it back to Sham, so 1st one is reports. So we have programmable reports. If you go to analyze reports, there are multiple reports that you see here, right? Inventory performance, switch performance. So let's say every Monday you want to see a report of your switch performance, then you can schedule it.01:16:41And you can also compare two reports. So let's say between last Monday and this Monday or Monday you downloaded a report and something broke, but you had a report from Friday. So you can compare the two reports to also understand what has changed.01:17:00Then we come to anomalies. Okay I should have said three things before I pass it back to but we come to anomalies. This is very common for anything on Nexus dashboard, right? If there is a deviation from your baseline in terms of drops, congestion, anything, then an anomaly is triggered. It'll tell you what cost it, what is the impact, and how to.01:17:15Fix it. We also have correlation in anomalies. Let's say your interface is down and that is causing your OSPF for BGP to go down. Nexus dashboard will also show that as a correlated anomaly. And now the last thing we speak about life protect. This is one of the features that a lot of customers are looking forward to using.01:17:36It's out now. What this does is usually when there is an advisory released right, all, all of us, we have this problem as to, you know, oh, now I have to schedule a downtime, you know, look at the fix, apply a patch. Now, all of that can be avoided all you with this live protect feature, as you can see on the nexus.01:17:56Dashboard we now have this card for advisory. It is showing you how many devices are impacted. So you can check the details of the impacted device. You can enable life protect and also if there is a fix available, right?01:18:15If at all there is an advisory that can be mitigated, then the system will provide something called as the signed compensating control which happens through a Tetra policy. But we basically what it does is how can you protect your fabric from that advisory or the impact of the advisory without doing a reload? So you.01:18:29We have three modes, protect mode, monitor and disable. So protect mode is when it will actively block the threat. Monitor mode is it's not blocking but it's logging any attempts to exploit that advisory and disable this we are not applying any policy. So this is one feature for which a lot of customers are upgrading their nexus dashboard to avail this.01:18:49Feature. If you want to know more about it, do let us know. But with that being said, I will pass pass it on to Sham for Ndy sand visibility. Over to you Sham.01:19:09Thank you Pradeepa for sharing those insights. Pretty insightful. Let me quickly the last 8 min try to share some of the visibility aspects for the sand side of the house. Let me share my screen.01:19:21Alright, let me move the slide. So we already talked about SAN insights, part of the telemetry which is equivalent to the nexus dashboard telemetry for the land fabrics. Now, a quick.01:19:37Overview of what we do part of this, right? Now, SAN analytics is the technical name to get this feature leveraged, which is part of the MDS switches. Now, what do we do is we monitor 80 different parts.01:19:52Parameters for every flow.01:20:12Now, what is the flow, right? Because a lot of us are aware of how a land flow looks like, but a sand flow when we say flow is every time the server will initiate the initiator, right? So that's the I and then it goes to the target, which is the front end.01:20:14Storage ports of the array, be it from NetApp, EMC, IBM, so that's the target. And then behind that, there'll be a lun. If it's a Scsi protocol or a name space if it's a NVMe protocol. So I, ITL, right? Initiate a target LUN or namespace. So this is the flow, initiator to.01:20:34Target and behind the target is gonna be a storage block which can be a learn or a name space.01:20:54For every flow, we capture 80 parameters and then we give insights in terms of what do, how can you take actions accordingly, right? So that's what we primarily do on San insights. I'll also show a few other data on the demo but you know given the time I wanna just ensure.01:21:00That all of you understand how it works in the backend.01:21:20In terms of visibility, we also already looked at dashboard. I'm gonna skip that, but when you go to each of these switches, we're able to look at the transceiver, the, the transmission, the receiver aspects along with the other parameters that we talked about, all individually and this gives you a lot of information as well.01:21:24We're also able to get something called as condition analysis.01:21:44So what is congestion analysis? I did give you a brief about something called as a slow drain. This usually happens in the sand world where initiator and target because of the mismatch of the capabilities, the size of packets that goes from either of it can create a congestion at one end. It can be.01:21:48Target site, meaning the storage array is congested or it can be the initiator side, right? Example, a lot of times we have servers with hpas of 32 gig, 64 gig, you know, a lot of options are there, but what if the target side is eight gig?01:22:08Right? And then what is the array be it brocade or MDS? It has only eight gig ports. So there's a mismatch of 6048 gig and then 16 gig and and what is the intelligence to reduce the packet flow, right? So there's none. So what happens is eventually one side gets congested and we're able to give you that metrics of.01:22:25How we could visualize the congestion. There's also another feature to mitigate this congestion. So that is called DIRL. I'll also show you the mechanism how we do that.01:22:45Now, in terms of visibility, we also have the insights into the compute and which is the initiator because we can integrate with the Cisco compute solution. We can also integrate to the storage leading vendors in terms of understanding how the target is functioning and of course MDS is ours and then we're able to.01:22:56To look at a lot of metrics in it and give you the end to end visibility.01:23:16So what I would also like to show is before I get into the next segment, I want to quickly show you where is SAN insights. Now, again, SAN Insights is a premier feature, just like nexus, which is you have essential Advantage premier. SAN insights or SAN analytics on the MBS front is a premier feature. Now, on the.01:23:19Nexus dashboard, you can enable that under the system settings feature management sign insights. When you do that under the analysis, under the analyze tab, you get sign insights. Now, when you go to the San insights, so my MDS is older hardware and I don't have San insights enabled, but you can see the metrics. Now I want to quickly.01:23:40Show you some information in terms of.01:24:00You know how things work on the backend, right? For that, I'm gonna quickly, you know, help you all understand how the transaction between the initiator and the, and the storage works. So, you know, let's call this as initiator and this is the target. So initiator is your server and target oops.01:24:04Yeah, let me annotate on the screenshot so that it's gonna hopefully remain. So there are a few metrics, as I said, close to 80 metrics, 80 metrics that have been observed, but also there are three important metrics that we observe part of San insights, which comes from the MDS switches. Eventually, that's pre packaged and then eventually analyzed.01:24:30And correlated to give you from the Nexus dashboard standpoint. So let's see. I hope now it stays.01:24:50So I was talking about initiator and I was also talking about the targets. Now, and, and also behind the target, you have the lunch space. It can be lun or it can be what people call as name space if it's NVMe protocol. Now you can see that there are two protocols available in San insights.01:25:00And NVMe, those are the two.01:25:19You know, ways that SAN world works. Now, always, we all know that in the sand world initiator initiates the traffic. So 1st initiator of the server is gonna talk to the target and it's gonna tell I I need this information. Now we all know what's in between, that's our SAN, which is basically.01:25:22Basically our MDS that's facilitating all of it, right? The zone zone set is all gonna be here that's facilitating it. Now, this th the target is gonna tell yes, I can provide you that and then eventually, the, the initiator is gonna respond saying that yes I can take it. And then the target is gonna give that particular.01:25:42Done and all the information and the entire process is acknowledged. So this whole process.01:26:02Is what is called as a ECT. It's the completion time of the entire transaction. And we also have another important metric, how long the storage took to respond, which is called as data access latency DAL, and also how long the host took to respond, which is HRL.01:26:08So these things help us mitigate and help us eventually.01:26:28Look at the SAN fabric and troubleshoot much better. So, higher ECT means there is a latency or condition somewhere, and along with that, if it's a higher DAL, then the issues on the storage side. If HRL is higher, then eventually the issues in the initiator side. If none of that is changing, then probably the issues in the SAN side.01:26:35Right, so this is where we're able to look at how.01:26:54Informations and how certain things can be affecting. So, I also know that we're over at time. I'm gonna quickly take two more minutes and then I'm gonna wrap it up, wrap up the session. In terms of visibility, we also have a few other information which I would like to show you. We talked about congestion.01:26:58Which you can see from the MDS fabric. Now we also have something called as, so of course congestion you're able to see the slow ten and we also have something called as DIRL. This is a technique from Nexus dashboard where we can limit the transfer at one side and streamline the packets so that there's NO congestion. So this is very powerful and.01:27:18And this runs every 24 h and also you can run on demand and right now I don't have any DIRL limitations, but if it does, then we will see all of that information. And last part of the SAN I would want to talk about is host path redundancy. We all know that initiated target will have multiple paths to reach, but is it on the right ways.01:27:38Then is there a port condition, is there a a drop somewhere? All of that you can't visualize it and run it all the time. So this gives you the ability to run that and give you that information so that you could be more proactive and prepared.01:27:58Having said that, you know, that concludes the sand part and also let's also look at few other advanced features that's part of the, the overall nexus dashboard arsenal. The 1st part is the multi site. So as you all know that a lot of our customers don't want thousand leaves in one single fabric.01:28:12Because it's one control plane, one overlay, it's gonna be a problem. So a lot of our customers would like to segment it, hundred switches, 200 switches. So when you segment it, then it becomes individual control plane and overlay, but how can we connect all of it and make sure it happens, right? That's a multi site solution and what facilitates this is border gateway.01:28:30That is ability, you know, that's basically connecting to the spine switches and connects to the DCI, which eventually bridges out of the conversation.01:28:50The next one is the multicast, so for all all the multicast traffic, we have very simplified format to make that happen with Nexus dashboard, leveraging the templates where you could define this on the VRF level VLAN level, switch level, and push that configuration. And basically this is TRM, right? Where you don't need the hairbending where.01:28:59Where you need the multicast to work. The 3rd is the macsec. As you all know, it's an encryption protocol and we can enable this on the fabric.01:29:19On every port, and this is line rate, meaning every port is encrypted without performance loss, right? That's extremely powerful, and Nexus dashboard helps you facilitate the configuration much easier. And the last three features we also help you integrate the layer four layer seven, just like in the ACI world, we could do that.01:29:29More than NXOS fabrics and we are able to facilitate that within the fabric outside across the multiple site. It can be a load balancer or a firewall, all of that is possible. The says 2nd part on this slide is the GPO. This is equivalent to folks who are working on ACI ESG, where we are able to segment workload based on IP, Mac.01:29:49As well as virtual machine attributes to group them and eventually this is the one that's facilitating conversation between VXLAN fabric, including ACI, which is IVXLAN to the standard VXLAN, right? So Gpo is the way that we segregate and micro segment east west traffic. And finally campus VXLAN VPN we saw some questions.01:30:09Where, what are the other switches that we support? Catalyst switches especially the 9000 series. A lot of our customers on the enterprise ask us, can we implement a VXLAN fabric on the catalyst for the enterprise? We we can do that all from Nexus dashboard. These are some of the advanced features, but, you know, that's some.01:30:29Something to leave you with where we could automate a lot of these for you, be it basic, mid day zero, day one, day two, all the way to advanced features. With that said, you know, in terms of conclusion, as already mentioned in the chat, you could always reach us reach out to us. We also have the recordings that we have provided the link here and you.01:30:49We will have that with you as well and very important, we also would like to know how this session was useful for the time that you spent with us. So absolutely you will get a feedback link or the feedback right after this after the session, would highly encourage all of you to provide feedback. But with that, with that said I'm gonna also pass.01:31:09To Pratibha to wrap up and add something that I might have missed.01:31:29Thank you so much, I think we're all good. So when we end this webinar, you will get the webinar survey. Please do take take it up and give us your valuable feedback and we have answered all the questions that you've got on the Q and A If you still have something or when you go back read about it, you come back with more questions. Join us on tube.01:31:33Wednesday, 10:00 A.M. IST, we will bring all our experts on the floor. All of us experts will be on the floor to address your questions live. Thank you so much for staying 5 min extra. Hope the 90 min was useful for all of you. Have a good day ahead.01:31:52
About this Video
Author : Cisco Customer Success
Uploaded : May 22, 2026
AI summary
Explore Cisco Nexus Dashboard Controller to design, automate and operate LAN and SAN fabrics. The recording shows Fabric Builder and Fabric Designer for best-practice VXLAN EVPN planning, Zero Touch Deployment for rapid switch onboarding, VPC pairing with Virtual Peer Link, and policy-driven VRF and network rollout. See image, patch and EPLD upgrades with group-based orchestration, fast RMA replacement, configuration compliance, backup and restore, and RBAC. Gain visibility with Connectivity Analysis, Topology, Endpoint Locator, vCenter integration, Reports, Anomalies and Life Protect advisory mitigation. For SAN, review zoning workflows, integrations, SAN Insights analytics, congestion analysis and DIRL-based mitigation, plus host path redundancy.
- Learn to enable Zero Touch Deployment and accelerate switch onboarding.
- Understand how to deploy image policies for safe, non-disruptive upgrades.
- See how to activate Life Protect to mitigate advisories without reloads.
Internal error
No related resources available.