Unified Edge Foundations Part 1

Hello, and welcome to today's expert insights series.00:11Where our topic is going to be unified Edge Foundations, part one, and with us today, we have customer success specialist. 1st one is Brian Morrissey.00:17And joining Brian will be Don Broke. Now we will be using Slido. Hopefully you can join us@Slido.com.00:34In the Webex clients, if you're able to use that, it's replaced Q and A, and you maybe able to change to Slido in the bottom right hand corner. Alternatively, if you have a.00:54Mobile device or a tablet, you can use the QR code on your screen that will take you to the join page where you can enter the passcode edge 1EDGE number one.01:10And the passcode will be the same edge one, EDGE number one. So hopefully you can get tuned in with us there. Now for those that are able to get joined.01:29We're keeping the joint information on the screen there along with the QR code. If you are able to join us there, we've got a question for you. We'd like to know the approximate scale of any edge or remote sites.01:50That you're looking to refresh or otherwise modernize coming up over the next twelve to 24 months. Now the results here will be for the knowledge and betterment.02:08Of our expert insight series along with the presenters here so that they can incorporate that valuable valuable information into our presentations. Now at this point.02:24I'm going to hand over to ball to good friend Brian Morsey and let him take it away to our presentation. Over to you Brian. All right, thanks Joe.02:45Alright, just want to verify you can see the PowerPoint.03:05Yep, I see your overview. Perfect. So whether you're in the initial stages of planning or have a large edge modernization project currently going on and this two part series will cover Cisco unified Edges solution. To start out with and just give an overview, AI at the edge.03:09Which is empowering organizations to respond proactively to evolving business needs and is fueling enterprise growth in infrastructure.03:29The shift is being driven by industry specific use cases that demand real time processing and immediate insights. In retail, AI is optimizing drive through experiences, making them faster and more personalized. In manufacturing, AI provides asset visibility and control enhancing operational efficiency.03:37And predictive maintenance insights. Why is this happening? There's a few primary drivers pushing compute closer to the data source. The 1st is data sovereignty. Many industries require data to remain within specific boundaries for compliance and security reasons. Processing at the edge helps ensure this.03:57Another driver's latency considerations for real time applications like critical health care, every millisecond counts processing at the edge drastically reduces latency enabling instant responses. The final driver's bandwidth needs, processing large volumes of data locally at the edge reduces the strain on network.04:16Work bandwidth, making operations more efficient and cost effective by minimizing data transfer back to a central cloud.04:36These requirements are creating a shift away from the traditional centralized cloud model towards what we call edge AI. Cisco Unified Edge fully integrates those key technologies in a modular design to meet the distinctive needs of AI edge environments while assuring.04:43Performance agility and efficiency.05:03The modular systems designed for the next decades supporting multiple generations of compute and networking technology including the Cisco 8255 secure router while meeting strict power cooling, space, and acoustic constraints. So we'll start out by diving into the hardware before jumping into the soft.05:05Software and demo that Donald do in a little bit.05:25So here we have the front view of the unified edge XE nine three oh five chassis. You've got your power button and status LEDs for local health monitoring in the top left. There's five slots for compute or networking nodes and NFC tag for management via the inter site mobile app. You have redundant power supplies.05:33For grid power mode and you have redundant ecmcs for remote management and northbound networking.05:53If you're familiar with other Cisco UCS compute products, you can think of the ECMC as similar to a fabric interconnect handling both the management and data planes. What you see here is the edge chassis management controller or ECMC. This is the central.05:59Hub that handles both the control plane, how we manage the system, and the data plane, how traffic moves through it.06:19Now looking at the ports on the front panel, we have a dedicated one/ten gig management port. This is your out of band lane used strictly for controlling and monitoring the chassis via inter site. Over on the right, we have 21/ten/twenty five gig ports. These are your primarily your data uplinks, and then for.06:25For times when you're physically on at the edge site, there's also a USB C port for direct local management and troubleshooting.06:45It's also important to understand what's happening within the ECMC itself. Internally, the ECMC connects to a high bandwidth midplane using 05:25 gig switch ports. The ECMC talks directly to the management controller on each of the individual nodes ensuring that the management and data traffic can coexist on the midplane.06:54Without getting in each other's way.07:14As you can see in this block diagram, the ECMC has two processing planes. It has the 25 gig system on chip for the data plane and a one gig system on chip for the control plane. There's a 25 connect gig connectivity to each node and to the peer ECMC, and there's also onboard EMMC storage.07:20With 55 gigs available for things like firmware.07:40For image staging, the ECMC uses internal storage to cache inter site images. You can optionally order a 75 gig M two drive and if both are present in the ecmcs, it becomes the primary image cache location. Intersight's device connector is gonna keep that cache automatically synchronized because.07:49Between both of the ecmcs.08:07One important note, the external USB devices can copy to the M two or EMMC, but they will not be mounted as virtual media for the compute nodes themselves. So now we'll move on to the compute nodes and we'll take a look at a couple of the different configurations.08:09Is to know about.08:29The 1st one we'll take a look at is the storage optimized node. The storage optimized node is built for capacity, pairs and Intel Zion six system on chip with up to 768 gigs across eight dims, has a GPU slot that supports a half height half length PCIe gen five card, also includes.08:33M to boot with optional hardware raid and up to 4E3S drives for up to a hundred and 20 tb of storage.08:53Networking is gonna come from a mid plane connected two by 25 gig Nick to the ecmcs plus there's also 21/ten gig RJ 45 warms on the front which will be enabled in a future firmware release.09:01The I/O optimized node trades one drive for an additional I/O slot, so instead of four drives, you get three up to 90 tb, but you do gain a second half height half length PCIe slot. Everything else CPU memory, GPU mix is gonna mirror the storage optimized node.09:19And then a quick note on the interfaces, the node includes Intel two by 25 gig mix that are connected to the ECMC mid plane and the 21/ten gig RJ 45 loms on the front of the node. The front pan panel ports are disabled at 1st customer ship and will get enabled through intersight in a future.09:41Release. However, if you want to enable it beforehand, there is a way to go into the BIOS options and enable those RJ 45 logs in the front.10:02In addition to the XE 130C, we've also recently announced the XE 150C compute node that supports one full height full length GPU per node, like the RTX Pro or the Nvidia L 40 S along with from accessible GPU display ports for your GPU intensive use cases like video stream analy.10:14Process or large language models.10:34Yeah, we'll go ahead and take a look at the internal networking and how you connect it to the outside world. Internally, each compute node is gonna connect to both ecmcs over the backplane.10:38At 25 gigs?10:57The two ecmcs are joined by a 25 gig switch to switch link. This design helps keep east west traffic local to the chassis. With regards to connectivity, there's two fundamental rules to follow. 1st management traffic requires its own dedicated.10:59Link. Second, the data plane utilizes either one or two uplinks from each ECMC. If you're gonna opt in for two links, they must be configured together in a port channel. As illustrated in this diagram, we have a one gig connection for our management links for each CMC along with a port channel from each ECMC.11:19That's gonna handle our data plane traffic.11:39For external storage, the recommended approach today is to connect storage to your external network switches, direct connect to the ECMC must be validated per storage vendor and disjoint layer two or direct connect will be coming in a future release.11:46And I'll pass it back to Joe for a minute to go into our second Slido question. Excellent. Thank you very much, Brian. And for our attendees, as you can see on the screen, we've got another Slido question, which is.12:06Regarding the virtualization or container stack that forms the current backbone of your plant edge footprint. We do have the Slido join information there on the left, the QR code for you, slido.com website.12:22Join code and passcode, but if you don't mind, if you could go and vote and tell us, the answer to that question there, that will greatly aid our future section, our future sessions.12:40For the export insight series I see we've got a vote coming in. So far we've had a choice of Microsoft hyper V We've got other options there, Red Hat, OpenShift, VMware, Linux.12:59Nutanix. We do have an option for other. If you've got something there that that you don't see listed. Right now we're at 5050 between hyper V and Linux. We'll leave that poll open for the next couple of minutes so that we can gather that valuable.13:14Information.13:34These vote and I'll turn it back over to Brian now. Over to you Brian. All right, thanks Joe. So now we're gonna shift into managing the platform, which brings us to Intersight. Unified Edge is gonna be managed through Cisco Intersight, which is.13:40One platform for all your UCS compute across data centers and edge locations. It helps consolidate tools, automates tasks, and delivers proactive insights, which ultimately free up IT resources and strengthen security.14:00A few foundational points before we start going into the demo later on. Unified Edge's inventory and configuration approach is similar to UCSX directs inter site managed mode implementation. So if you're familiar with that and inter site managed mode, the unified Edge configuration is pretty similar as.14:19As well. In addition, there's two firmware bundles similar to inter site management as well. So you have your infrastructure and server bundles and it's supported on both the inter site SaaS and the connected and private virtual appliances.14:39For communication with intersight, there's two types of device connectors in the picture. So each CMC runs one device connector in its control plane processor. This handles chassis and server discovery, inventory and ECMC networking and chassis configuration.14:57Each compute node also runs a device connector in its Cisco IMC, and this reaches inter site through the ECMC device connector proxy, and on the compute node, the device connector handles the server level inventory, profiles and firmware.15:14For onboarding the unified edge, it can be driven through the recovery console web UI or via NFC tags and the inter site mobile app. For the web UI, the ECMC is gonna default to DHCP on the management port and provide a host name via DHCP option 81.15:33If you don't have DHCP available on that particular subnet, it's gonna fall back to a link local address after 60 s. So in the case of static IP addresses, if you need to configure those, you can set your laptop to something on the 169-25-4254 subnet, and then from there you can.15:51Load up either ECMCA or ECMCB, which will be on the.one and.two re respectively.16:10Once you log into the recovery console, you can set the admin password optionally configure networking, and then grab the claim code and device ID to claim the chassis and intersight, which Don will show in the upcoming demo. Traditionally setting up new infrastructure.16:18Sure requires external external coordination and in the case of edge deployments, it also involves staging at a central site leading to delays and inconsistencies.16:37With the Cisco unified Edge and inter site mobile app deploying infrastructure at any location will simply require tapping the mobile phone on the Cisco unified Edge system. Limited to NO technical expertise is required to claim the device. So whether managing ten systems or 10000 organizations can achieve remote.16:47Put in rapid onboarding without requiring extensive onsite expertise.17:07In addition to NFC based claim, Intersight now also provides the ability to give temporary limited access to Intersight for users without single sign on accounts to simplify device onboarding. A user's access is bound to a specified time window and a specific role or operation. The guest user will be emailed a link to.17:16Log into Intersight and use a one time password for authentication.17:36And that guest user will have access based on the curated workflow and experience that's assigned to their particular role. This is useful if you want to give someone access remotely, so maybe you have some smart hands. You can provide them with guest access to Intersight where they can perform device claim task.17:40In regards to the setup of the unified edge, it's driven by profiles similar to other UCS platforms. The unified Edge profile combines the domain and chassis profile concepts into one, which we'll see in the upcoming demo. It defines chassis wide settings like your ECMC networking, so which ones are uplink port.18:05Parts, the roles, the VLANs, your QoS settings, your MTU. Additionally, any kind of power and thermal policies and traditional system management settings like LDAP and local users, as well as monitoring via SNMP and syslog.18:25The UCS server profile handles the compute side, much like inter site managed mode or standalone C series rack profiles. It covers server assignment, your BIOS, your boot order, your firmware, and your network configuration, so like how many vnics that you have defined, which VLANs and then the QoS settings for those.18:44In the upcoming demo, we'll show you what that looks like and that'll help set the basis for the upcoming part two webinar on fleet management. So with that, I'll pass it over to Don to go over the fundamentals of unified Edge within Intersight.19:05So, as Brian said, this will be a part one and we'll cover a little bit more of the bringing unified edge into Intersight and deploying workload on it. Next week we will expand things further with the management blueprints and all that fun new stuff.19:21So I'm gonna start off with a video. This will be the only video, everything else this will be in live lab, but I have a video displaying how to claim the device in intersight. So we will start off with that and then go from there. And I'm gonna talk you through.19:38This process. So starting off in Intersight, if you notice one thing I did want to call out on the left under the operate, you see what you're used to seeing servers chassis fabric interconnects, you've got Nutanix, you'll see Nutanix clusters in there.19:58All that sort of stuff. After we bring unified Edge into the, into the inter site, you'll see a new option there for unified edge. So let's go ahead and hop over to the device console for unified edge. And if you've done any inter site managed mode.20:15With traditional fabric interconnects and all this is gonna look pretty familiar to you.20:35So we'll go ahead and log in with the admin credentials that were added at the time of deployment and once you've kind of done the initial install and you'll come in directly to the device connector page. So right here, just like with your fabric interconnects, you're gonna have a device ID, you're gonna have a claim code. You'll see that the device.20:40This connector has reached out to find inter site, found intersight, and it's not claimed. So it's sitting there waiting for us to go ahead and claim that inner site. Some other things that you can see here at the device console, it's again, the device ID, the claim codes alright here.20:59If you go to settings, if you need to do proxy, you can set your proxy in here if you need to turn off the device connector for whatever reason, just like in in inter site managed mode, that device connector toggle is there for you to do, here you can toggle to enable proxy.21:20See and when you do that, you can put your proxy information.21:39Your certificate managers right here and then you can check your connection to Intersight depending on which inter site you would need whether it's Americas or EMEA. Here in system information, you're gonna get some basic info on the hardware itself. So you'll get your host names that are out of the box.21:43To see we have the ECMC for both the A side and the B side is we've got our configured management IPs, we've got our serial numbers, the current firmware that's installed. Under inventory, you'll see we have any servers that are installed in there will show up here.22:00We've got the secure router units installed, those will show up there. We get some chassis information here, basically, you know, the name, the serial number. And then this is that internal storage. Let's go back to that real quick. That's that internal storage that Brian mentioned.22:19Earlier, that is there for holding onto any firmware that you might be using to do any firmware upgrades on the system. So this is where you can kind of see your storage utilization for that if you need to.22:33So let's go back to the device connector, well real quick, if you needed to generate a text for bundle from the device, the device console, you do that there. Let's go claim this guy in intersight. So we're gonna go to our targets and we'll claim a new target.22:52From here, we're gonna see this new option for Cisco Unified Edge. That's what we're gonna wanna select and then go ahead and click start and we'll go ahead and grab that device ID and paste it over in inner site.23:09And then we'll go ahead and grab that claim code as well and paste that. Site information's there, we could go ahead and put a location in here but I don't want to get too much into that because that you're gonna learn a whole lot more about next week when we talk about.23:26For now we will go ahead and click claim now, and it claims our devices in intersight. Now, once that is claimed.23:50We should see up here now we have this unified edge option there. So just like with the regular intersight, we can click on there, get some information. Right now it's not seeing a whole lot because we just claimed it and it's still at this point hasn't gone through. It's.24:06Discovery, if you notice up here, we've got some tasks that are running, so it's doing a discovery of the inventory right now. So it's waiting for that inventory completion.24:26You just kind of get information about what is physically installed in this particular unified edge chassis. I'm gonna skip forward through this because this does take a few minutes as anyone who's claimed new inter site managed mode devices before and the discovery.24:38Can take some time. So it's gonna go through and wait for BIOS posts, server, and it's doing this so it can pull the inventory of the server itself.24:58Things like its network adapters, memory, all that stuff, physical stuff installed in the server, it's gonna pull the, this information directly from the server by booting it and reading it from the device connector that's installed there directly on the server.25:18It's waiting for the server inventory to complete. Once that's done, it powers the server back off, and then device is now claimed in intersight. So let me stop sharing this video and let's jump directly into that inter site lab. So now here we are in that inter site, we're gonna go look at that.25:35Name server that we just claimed. If you look here under operate now we do have this unified edge, and then under configure, you should see if we go look at profiles, you should see now this unified edge.25:55Profiles tab up here in addition to your chassis profile, your domain profile and your server profile. So, let's walk through what we can see in the hardware here in intersight, and then we'll take a look at the the different profile options that we have for the unified edge itself.26:14And for the servers.26:34So we come here into operate and we click on. This is device right here. If we click on that device, we're gonna be able to see a whole lot of information just like we could or would with any other server that we bring or chassis or or fis that we bring into Intersight. We have our.26:36Health way over right here.26:57So it's gonna give you health of the PSUs, you've got your front and rear view. There's not really much to see on the rear view, just a bunch of bands, view, you can see we have those five slots, four of them are empty and one of them is holding a single used unified edge UCSXE server.26:58Down here we'll get some basic information on the chassis itself, the serial number, model, contract information, if you have any tags that you use, you could set those here under inventory. This is gonna look a lot similar to the.27:20The the device console itself, whenever we clicked through the inventory on it, there was gonna be all of our fan modules, their state power is of course gonna be our PSUs.27:38And their state and then service is gonna be a list of servers. We do not have the, the networking module installed in here, so it doesn't show up here, but if we did, we would see that here. Notice you also have your decommission tabs, so if you've gone through and decommissioned any servers.27:53They would show up here under this service link. Unified edge profile is gonna show us the profile that is currently assigned to this unified edge device and the policies that we're using for that, and we'll dive into that here in just a few minutes.28:12Topology is gonna give us basic topology of the device, much like what you would see in with your fabric interconnects with these are two ecmcs, and they are providing.28:30Networking to this server down here specifically cause it's the only server we got. And under metrics, you can see we'll have some environmental metrics, some networking metrics, any errors.28:44So this might be something you want to look at if you're noticing or getting complaints of any weird potential networking issues or at the edge, environmental is a lot harder to control than in a data center. So you might wanna monitor any.29:00Temperatures or anything like that, your fan speeds through this console here just to kind of keep an eye on things. Okay, so let's go take a look at the.29:15Unified edge profile. So this profile is for the chassis itself, so think of this as kind of like a combination of your domain profile and your chassis profile with an IMM domain.29:31So I have a profile already created and assigned. I'm gonna walk through creating one right now so you can kind of see it. But I'm gonna go ahead and keep the one that I have assigned out because if I unassigned and reassigned it just take a lot of time and we don't want to do that, but I'm gonna walk you through creating a profile.29:46And it's gonna look real similar to creating any other profile that you've done in intersight. So we'll name this demo, and you'll see some of the options we have now. You can assign it now if you have an available unifi.30:03Right edge or if you are just staging for an upcoming delivery, you can assign it later. We'll just stick with a sign later.30:22Our and thermal policy. So there are a couple of things to note with the these policies for the unified edge chassis and Brian touched on it when he was going through his presentation. I'll create a policy real quick and show you.30:32Here it's where you're gonna set your power redundancy, in a normal UCS IMM domain, you would be able to choose from grid, not redundant in plus one in plus two.30:51With unified edge, the only two supported options at this time are either grid or non redundant. So if you try and do an N plus one or n plus two, when you go to deploy this profile onto the hardware, it will fail. So make sure when you're doing this, if you have redundant power, you would select grid and just make sure you have.31:06Have your source of power coming from two different systems plugged into that device or you could do non redundant. For power allocation lots, this is where you can.31:26Cap the amount of power that the devices are consuming. However, with unified Edge at this time, a power allocation of zero, which is basically NO cap is all that's supported right now. Go ahead and create that policy.31:44For the thermal policy, this is basically how do we want our fans to behave? So we'll create a thermal policy and here our options are gonna be balanced.32:01Low power, high power max power acoustic or maximum cooling. If you do not set a thermal policy because it is not a required policy, the unified edge will default to acoustic. Acoustic basically just means it's shooting for that balance of.32:18Of cooling, but not being not running full speeds, not being too loud, that might be important depending on the the setting that this device might be installed and you know if it's in a retail or a branch or something like that and it's in a closet that might be close to customer ears.32:34You know, we're gonna try and keep the noise as as low as possible for for this device. So that's what it will default to if you do not set a thermal policy. So that's the two just things I wanted to bring to your attention on these two policies.32:52Next is where it gets real fun. This is our switch configuration, so our VLAN configuration and our port configuration. So if we were gonna create a VLAN policy, it looks very similar to creating a VLAN policy for.33:08For inter site managed mode, basically you just come in, give it a name, Notice target platform is different from UCS domain, so these, this is not a one of those.33:27Policies that can be shared between different types of profiles and click next. And then you can add your VLANs, you can do your prefix, and then your VLAN IDs, you can do individual VLANs or you can do VLAN.33:42Ranges so we could do like 8093 and and that will list out all of those VLANs there. If you need to make one of these native.34:00You would do that by selecting whichever one and then selecting, well you're gonna select it, just select native ID and then put that VLAN ID there if you needed to just to set a native at all those VLANs.34:14Port configuration. So this is how we're gonna tell our ports they're gonna be. So let's go ahead and create a new port policy.34:33And just like with Fabric interconnects when we're doing a domain profile, we have a few options here. So if we're doing individual ports, we would set those here just by clicking on this port.34:53For that port, whichever one and clicking configure and set that port rolling. It's either going to be unconfigured or the internet uplink. As Brian mentioned in his presentation, if you're going to use both of these ports, then that has to be done in a port channel. To do a port.35:07Channel we would just click on this tab here and click create port channel. It's gonna be an even uplift port channel, you can give it a port channel ID, and you could choose speed if you don't want to do auto, if you need this to hard code it to ten gig or 25 gig, do that.35:27The link aggregation policy, this is basically LACP. So I'm not gonna go through showing you that because there's nothing real fancy there. It's just, are you doing LACP fast LACP normal, are you doing anything crazy with your spend?35:46Anything like that, I'm just gonna go ahead and select the policy we've already got. Okay, so one little thing I just noticed with the port channel ID, it has to be either port channel one or port channel two. It does not like anything other than that. So we'll just go with port channel one.36:01And here we're gonna select which ports are gonna be in that port channel. There's only two, so it makes that decision pretty easily, and we'll click save. So now we've got our internet uplink port channel. It's port channel, It's one and two, and it is port channel identificator.36:20Identifier as port channel one, save and we can do the same thing on B B, we can go ahead and do the policy that we just created for the VLANs because as Brian mentioned right now, disjoint layer two is still feature.36:37That's coming in the future. So we'll just lect there. With the port configuration, you can either reuse or you can do a separate port policy for your B side. Typically I'd recommend doing a separate port policy for your B side, just in case.36:56You want a little bit more control whenever you're doing some maintenance and you take that one side at a time, make a change to the policy. You make a change to a policy that is a reboot requiring change and you accidentally make a mistake, then you're having to reboot both sides.37:13More more times than one. So I like having a separate policy for each, but for the demo purposes I'm just gonna go ahead and select the same policy that we just created. And then down here we've got our switch configuration. So that's gonna be our switch control policy and our system QoS, and NO system QoS is the only.37:29Policy so far that is a required policy, and that's that's the same for imm. So that's, that's common amongst the the platforms. So we'll go ahead and select the policy and I will show you a couple of things with this one.37:48Down here, notice unlike with IMM, there is NO fiber channel.38:11So fiber channel is on by default on the IMM domain profile configuration. It doesn't exist here on the unified edge profile. So that's just something to note. Much like with IMM, the best effort is default with a cost of.38:15Me and the weight of five, you can adjust this as you need. If you are using other QoS policies.38:35Platin gold silver bronze, just like with IMM, you can enable those and adjust the cost value and the weight as you need depending on what application you're serving and what their, what their needs are. But by default, it's gonna have those disabled best effort enabled and NO fiber channel in the policy.38:43And switch control. So if we look at our switch control, basically, all that really does is set your jumbo frame settings. So by default, if you don't set a switch control policy, jumbo frames will be enabled.39:03So if you need to disable that for whatever reason, maybe your upstream switching is not set for jumbo frames or your ISP doesn't allow that whatever the case maybe, you would create a new policy here.39:23And just toggle the jumbo frames to off after that we'll click next and we'll get into our management configuration, which is all the fun stuff.39:41If your NTP provider, if you're doing syslogging, you would put that information here. Network connectivity is our fancy word for DNS, so you would use a, network connect network connectivity policy to define your preferred DNS servers for the system.39:57And then local user. So the local user policy allows a user to log on locally for troubleshooting potential if need be. So we'll show you what this looks like.40:16Give it a and here you would go ahead and click add a new user. You can toggle as enforce strong password on or off if you want. We recommend, of course, having that on.40:34You're gonna have a few roles that you can choose from. Read only, of course, is gonna be read only so that whoever used this policy would have just basically visibility into the system, this particular unified edge that we were applying this to, but not the ability to make any changes.40:52Or you could do admin or user. So we'll give it a username and give it a user password.41:12Create that policy and next we're ready sorry I'm not gonna set these because I mean, you guys have seen NTP, you've seen syslogging, you've seen network connectivity before, so not breaking any new ground there. Go ahead and click next.41:34And here, just like with any other profiles that you're doing, it'll give you a chance to review the settings that you've made. If you need to go back and change anything, you can. This one's ready to deploy because I selected deploy letter, the deployed later, the deploy option.41:53It's grayed out, but this profile is created at this point. So if I click close, that profile is now here, not assigned, but it's ready to go or when we have new hardware installed and claimed and we're ready to put a workload on it.42:10So that is creating a unified edge profile. If we go click on this profile that we have that is currently attached to the hardware we've been taking a look at, you'll see basically the configuration that we have for that. You can see the list of policies. You can take a look at that policy to see how it's configured.42:25If you need to, if you need to adjust that policy, you can do so by clicking the pencil icon here, it will bring you to that policy and then you can make any changes here as you need.42:44So that is creating the profile for the unified Edge specifically. Some of the options we can do here is on this particular profile we can deploy it. If it's not deployed, we can unassign it. If we need to actually move this profile off of a.43:00Of a unified edge chassis to apply another profile to it, whatever the case maybe, we can clone that policy, that profile, sorry. We can attach it to a template. That's another thing. If we go to our templates, we also now have unified edge profile templates. So if you're going to be deploying several of these that need to have light configuration.43:20You can just create the unified Edge profile template and then spin up multiple profiles off of this template and then apply them to the different devices as you need. So now let's go ahead and take a look at.43:40Putting a workload on one of these servers. If you go to the service section of your operate, you'll see the unified Edge servers will show up here just like any other servers.43:58You can note them by this model number UCSX E that tells me that that is the unified computing systems unified edge. And we click on that server just like with our our X series or M series or C series.44:14You can take a look at some more details on said server get its management IP, its serial number, take a look at its licensing, its firmware, HCL information, all that fun stuff.44:30If we take a look at the top in the health, just like with our other servers, we can take a look and see the health of our CPU. We could take a look at the dims. And so if we had a dimm that was having a.44:45That error and we could of course show red here so we could point that out and kind of walk remote hands through replacing that if need be. Some of the actions that we can do here on the hardware is we can power the system on and off, of course, hard reset it.45:01Shut down the operating system, we can turn on the locator LED. So if we needed to direct someone to this particular server, we could turn on that locator ID, that little light on the front of the server would start blanking and make NO, ok, that's the one I need to go and take a look at.45:17You can do a rediscover if you need to do a rediscover of the server for any maybe troubleshooting or anything like that, that option's there decommission. If the servers reached the end of its life and it needs to be removed, you can decommission it. You can do a security race. You can reboot the management controller itself for the server.45:34Setting in asset tags, user labels, you can enable and disable tunneled KVKVM from here PE two and then system logs. You can also do your install of your operating system, which we'll take a look at in minutes.45:51You can do your firmware upgrades from here, you can launch VKBM or tunnel VKVM if it's enabled and you're licensed for that. From here, you can start or stop your alarm depression, open attack case, so if you're having an issue and you need tax assistance, just like with our other systems, you can open attack case directly on this particular.46:09Hardware and go ahead and get that going. And that everything that comes with that so the connected tech means they're gonna be able to pull the logs.46:29For you, if arma's all included, all of the fun stuff that we get on our other systems is is available here as well. You can set your licensing tier. So if you are a mixture of of licenses.46:36Advantage and essentials. You can switch between the two if for some reason you want to run some systems on essentials license and others on advantage, you can set those at an individual level on the hardware. And then you can also collect the tech support bundle if you need.46:51Under inventory, it's gonna give you a list of all of the hardware installed in the server, so that's our CPU, that's our memory dims, that is our network adapters, storage controllers. We can take a look at the server profile that's associated with that server. We can take a look at.47:08At the HCL.47:28And like other systems we can kind of take a look at the topology and kind of see how things are connected. And of course we also have metrics. So that's the actions menu just briefly for the server hardware. Let's go ahead and let's put a server profile on this thing or let's Let's go and create one. I'm not gonna, like I said I'm not gonna.47:32Software profiles because I don't want to sit here and take everyone's time waiting for these things to booth, but we'll walk through creating a profile for a unified edge server. Those will actually be done over here under the server profiles.47:52So this is the server profile that is currently on that device. We're gonna create a new server profile, and I'm not gonna walk through every policy because a lot of them are very similar to what you're used to in IMM, and if you're not super familiar with IMM, we actually do have some EIS and expert insight series videos that we've done on IMM I have.48:11I recommend you to check those out.48:30So we'll go ahead and get started. We're creating a new server profile. So say we have a few options here. We've got UCS server standalone, so that's gonna be a C series server that's not connected to fabric interconnects. We've got our UCSFI attached servers, so that's gonna be either.48:33Email servers or X series servers that are connected to.48:53But now we have this UCS server unified edge option. When I click that, you're gonna give it a name, go ahead and get started.48:59So we can here select a specific server if one's available. If we were using resource pools, we could just pull a resource pool, we could sub define a specific chassis ID and slot number. We can assign by serial number or we can just go ahead and create the profile now.49:15And assign it. We'll go ahead and do that because once everybody already has a profile on it, so it's not available, so let's go ahead and assign later. And once I am, we're done walking through the profile creation, we'll go and take a look at a few things on that one that already has a profile on it.49:34So, our UUID assignment, so this is where we select which UUID pool we're gonna use and I'm not gonna walk through creating a UUID pool that's, you know, something that's been around for a long time. I'm just gonna go and select a UUID pool that is already available.49:53And select. And we've got all of the different settings that you're used to seeing when you create server profiles, our BIOS, our boot order, our policy virtual media policy. I'm not gonna show you a ton of these because there's nothing real new here.50:10I will go through the BIOS policy real quick though, to see what, if anything is different on a UE, a unified edge. So we'll go ahead and create policy and.50:29Get this. So down here we can do two different policy types. We can do generic or we could do model specific. Now, one thing to note that the.50:45Generic BIOS is really only compatible with the UCSBC and x series models. So using any unsupported model can cause configuration errors. There's a link to more information about that for model specific.51:02That is going to be compatible only for the C 845A model, which we are not working with here. So we're gonna stick with generic. We can use this Cisco predefined configuration. So we toggle that, that's gonna allow us to choose.51:20What type of workload the server is gonna run without having to go through and set individual by settings. So let's say this is an mate server cause it's an inmate server, and it's running Intel processor and the workload type. So.51:36This would be where you determine what sort of workload this server is gonna run. If it's gonna be a hyper V server or or an ESXi server, you would pick server virtualization, right? If you're doing some high performance computing, you would pick high performance computing.51:54You're doing databases, you've got different options here. I'm gonna go and say this is gonna be a virtualization server because we've only got one hardware service, so I might need it to run multiple different Vmiss or containers or whatever the case maybe, and click next.52:10Now this part takes a second for it to switch through because it is still gonna load all of those bio settings even though we've done that, that pre configured BIOS. This means this gives you the opportunity to come in and maybe make some individual tweaks that you might need to for individual bio settings.52:27Depending on the workload that you're gonna deploy, you might need to do that. I'm gonna just go ahead and click create here. Boot order, I'm not gonna worry about, so I mean this is basically selecting what your boot order's gonna be. These servers for.52:45Now we're gonna boot off their local storage. They have dedicated boot hard hardware for that scrub policy media policy I'm not gonna worry about any of that cause we're gonna show something later and all of this is not new. This is all stuff that is.53:04And around in inter site management for a while. Same thing here, you know, IMC access, the local user, we just talked about local user, we've already shown you that policy. Virtual KVM, this stuff is.53:24Been around for a while so I don't want to waste a lot of your time with it. So we'll go on to storage configuration, you'll click select policy here. Let's go ahead and create a storage policy. This is gonna be for your local storage.53:40We'll turn on our M.two raid configuration, so this is where we're basically configuring that boot rate for our OS to be installed on into boot two.54:02It's gonna default to this in store boot BD and depending on your configuration, you can do a raid one, raid two or this raid one and raid two. I'm gonna stick with the defaults, just do a RAID one.54:13Next, and then the LAN connectivity policy. So this is where we define how we're connecting to the network. If you take a look at this policy that I already have created.54:32It's gonna use a few other policies. It's gonna use a QoS policy. It's gonna use a network group policy for both the A side and the B side for our connectivity. So let's go ahead and create one of those.54:45And again, notice the unified edge is already pre selected for us at this point. So this can be zero, A, whatever you wanna name it.55:09You will choose your side. It's either gonna be your A side or your B side. So we're gonna do this in the A side for our internet network group policy, so go look at one that's specific for unified edge.55:27Just kind of take a look at that. That's basically what VLANs, right? You know what things are we allowing on this nick? Is anything a native VLAN? Are we doing Q and Q tunneling? I'm not gonna go through and create this cause this is not really groundbreaking stuff. It's just basically where you define the VLANs.55:43That are gonna be put on this virtual nick. So I'm just gonna go ahead and block the one that's already existing and then QoS policy.56:00Same thing, we'll take a look at this one that's already configured, and it's basically what is my NTVU on this device gonna be? Am I doing jumbo frames? If so, that might, that would be 9000. If not, it'd be 1500.56:19Am I doing your rate limiting? What's my priority? I only have one priority of enabled, it's best effort, so gonna default to best effort. So I'm just gonna go ahead and select this policy. Excuse me?56:34And go ahead and add that. Now that's only the A side we probably wanted either that on our b side too, right? So we'll do eth one that's default to the B side because we already assigned a vnic on our A side.56:49Go ahead and select the policies. We'll just use the same policies So that's our A side.57:05Internet switch adapter and our beset internet adapter, we can go ahead and click create. And that's the last policy for the server profile. Once you hit the next button it goes to that same page where you can review all of the changes that you've made or you, all of the, you know, the settings that you've made.57:21Make any changes if you need to, and then that server profile is now created and ready to go. There is NO server for me to assign it to, but it's sitting there ready for a new server to come along for me to then assign.57:38And go ahead and bring that up. But I have this one right here that is already assigned to that hardware and so if I click on it, you see a lot of the same information that you'd see on any other server profile, get some information about the hardware itself.57:54You'll see the configuration for that server or that server profile, if you click on server, that's gonna kind of pull information over here from the operate tab, and we can get some information about the hardware and then inventory. Of course, we've already looked at. That's gonna be the inventory of the physical hardware there.58:12Okay, so we've gone through profiles for the server itself, we've gone through profiles for the unified edge chassis itself. We've looked at the inventory in the chassis view, we've looked at the inventory and the server view.58:30Let's talk a little bit about operating system install. One of the new things that we are introducing with unified Edge is the ability to store.58:51OS install media in a, inter site hosted location. So it because these unified edge servers are going out.59:09To, you know, theoretically multiple different different locations amongst a wide geographical area, you may not always have the ability to either host an ISO and install media locally to that server or maybe it's just you could but it's so far away that doesn't make sense because you're.59:25You're transferring, you know, ten gigs over a WAN link, right? So, we are now making it to where you can actually upload these install medias to an inter site hosted.59:44Repository and then you can use those for your OS install. So to do that, if you would click here on add OS image link, you could, of course, if you're using different orgs, you could assign it to different orgs. Then you have the typical options that you've always had NFS, SIP show HTTP.59:59Sip share or HTTP HTTPS, but we now also have intersite hosted. So if you click on intersite hosted, you can browse to any ISO that you might have what I've got on here.01:00:18I don't have any ISOs on here, but I mean you've uploaded something new to places before, so you would browse for the install media, click next, you would give it details.01:00:37Details would be, let me close this and show you what details are gonna look like. So if I click on this Ubuntu, these are the details. You're gonna give it the a name, what vendor is it from? So is it is it Ubuntu? Is it Red Hat? Is it VMware, is it windows? What's the version?01:00:52So you'll give it the version of the install. For instance, this one is Ubuntu 2604, default to intersight hosted the image size, of course, it's gonna be dependent on what you're uploading. So once you've uploaded that, those are now stored in Intersight for you to use NO matter where your.01:01:07For unified edge devices are. So if we were to go and do an operating system install, go to operate, look at this UCS server, if we click right here on install operating system.01:01:27So we would start the install operating system workflow, go ahead and select what operating system we wanted to install. So here it's gonna have all of the install media that I have access to, either in local repositories or in those inter site hosted repository.01:01:47If I needed to add a an OS image link that's not included here, I could do that directly from here just by clicking that and it brings us to this, install operating system media upload option there. So let's say I'm gonna put red Hat on this server.01:02:06Next, and this is just like really any other OS install, which you choose either Cisco custom or embedded. What does that mean? So Cisco is, it's going to default to a validated template for the operating system.01:02:26That configuration file that it would use is right here. You could actually click on this and see the information it would use for the actual installation. If you had your own custom configuration file, you would click on custom and then you would either put that custom file in a repository.01:02:44Or upload it from your local machine.01:03:04And embedded is gonna be when that file is actually embedded in the operating system install media itself. So we'll go with Cisco custom or Cisco, sorry. And here we're gonna provide information like IP address of the operating system.01:03:06It's netmask gateway, preferred name server, what do I want its host name to be? And what do I want its password to be, I need to find this information so I can move on to the next.01:03:24Screen and show you, so bear with me just a second.01:03:42We'll just throw something in here. So.01:04:04Give it the name server, so this say it's gonna be rail.01:04:32The Branch, two oh one networking device. So, that is which Nick are we gonna use for this install view the blank, it'll just find.01:04:45Default one and use that we give it a password.01:05:05Click next. So here we're gonna choose our server configuration utility. This, you can again have either inter site hosted or if you have one specifically that you have already on your local repository.01:05:11Let me do that. So, I'm gonna go ahead and grab, I believe this is the firmware that we have, so I'm gonna grab this one. And what the server configuration utility does is it, it's the, it's the initial ISO that we use to boot the server up to, and it basically facilitates the automated operation.01:05:28System install from there next and so we have our installation target. What do I mean by that? So that is where are we actually installing this operating system on? Is it be on a local disk? Is it gonna be on a fiber channel or iSCSI connection? We're gonna do local disk.01:05:48And it is gonna be using that in store boot VD drive that we had created in the profile. And then once you have all this information done, you are ready to go ahead and click install and kick that off.01:06:07I'm not gonna do that because I already have Red Hat installed on here and it takes time. I mean it's you know there's several reboots involved, it's an operating system installed non instant, so I don't want to take everyone's time showing that I will do from here.01:06:23Is go ahead back to that server and show you a few things here that we can look at. We kind of touched on HCL. This is super important. See I've got some.01:06:39Things I might want to take a look at on this server to see, why they're not compatible May maybe i'm running the wrong driver for this particular device. That's something I want to dive into, but this HCL is gonna show us our our hardware, software, and compliance. So are we running the correct.01:06:55Drivers for the OS version and we running the correct OS version for the firmware of the hardware that we got it on just to make sure that everything's in a good, tested valid set up that we can put our production environment on there.01:07:15And on that with with you know without any issues and I believe that is pretty much the end of my demo. The only other thing I could show you is how to do firmware upgrades from here.01:07:33So we are still here on the physical server itself. So if we go to actions and we go to upgrade firmware, we start this process, here we'll select the server. We could do multiple at a time as long as it's the same hardware version.01:07:50So it'd be the same model here, trying this to expand, there we go, the same server model there and we're going to the same firmware version.01:08:07If you, if any of that is different, you have to do that in different steps. So we'll go ahead and click next. Here it's gonna pull a list of available firmwares. These are hosted directly on Intersight, so you don't have to go and grab these, you can get them directly from intersight.01:08:23This little blue thumbs up, that is the suggested release. If you've done any firmware upgrades on UCS manager or inter site management in the past and you've gone and gotten those firmware bundles from cisco.com, this is the equivalent of the gold star release that you would find here.01:08:39So you could select that or if you wanted to go higher for whatever reason, you could select whichever version you wanted to and click next. Then it is going to give you the summary page, it's gonna give you the version that you're going to.01:08:58Is it reboot required? Things like that just kind of give you some information on what you, where you're at and what you're moving to, and then you could click upgrade and it would fire off the upgrade, the firmware upgrade process on the server. We go to unified Edge itself.01:09:13Click on it, we have configure firmware upgrade here, very similar process. It's gonna select the chassis that we're looking at.01:09:34And we click next and we would then select a version that we wanted to upgrade to next, and here we would specify a specific timeout if, if they request.01:09:50Takes too long, the default's gonna be 30 min, and the upgrade process itself would time out and user acknowledgement if we decided that, that would be the upgrade request will sit there and wait to before upgrading each ECMC.01:10:05For the user to come in and acknowledge to go ahead and restart that particular side, A side or B side ECMC for the firmware upgrade. And here is that summary page again where it's gonna show you the version you're upgrading to, the version that you're at.01:10:22And this is your, you know, your last step before, you know, either click configure and you go forward or you hold off and maybe wait for a maintenance window or different maintenance window or whatever the case maybe, but this is your point of over time once you click configure on that and go ahead and click cancel there.01:10:42And that is all that I have to show you. This week again was part one, so it's a bit more of of kind of just the the initial configuration, our initial claiming and setup next week we're gonna dive into really cool stuff like blueprints and location.01:11:01Patients and fleet management.01:11:21So stay tuned for that. It's gonna be real good for now. Joe I believe I'm turning it back over to you. And as you mentioned, firmware is a great segue into our part two webinar next week where we're gonna build directly on the foundations we covered today and show you how they apply to fleet management at scale.01:11:24Just to give everyone a quick preview, one of the most exciting additions to our fleet management strategy is the introduction of blueprints. You can think of these as templates for your edge deployments. They come preloaded with best practice configurations, specifically designed for your workloads, taking the guesswork out of the setup.01:11:43The blueprint based approach offers three advantages for anyone managing edge environments. Number one is consistency at scale, it ensures that whether you're deploying ten sites or a thousand, every single one is gonna be configured dynamically. Number two is the golden configuration. It allows you to establish a fleet wide standard or golden.01:12:02In config that ensures that every device meets your specific security and performance requirements and finally zero touch provisioning. So this is the ultimate goal with zero touch provisioning. You can ship hardware to a site plug it in and let the blueprint handle the rest automatically. So we'd highly encourage you all to join next week for part two to see.01:12:22Those blueprints in action and with that I'll pass it back to our moderator Joe to open up the floor for Q and A and share some final closing information.01:12:42Excellent. Thank you very much Brian, and thank you Don, for that excellent demonstration that you did for us. For our audience, we're going to move to a time where we can do some Q and A as Brian mentioned, and I'm going to post some links for you that.01:12:52Should turn out to be handy. I'll have links to.01:13:12Not only our Cisco community, but also for the expert insights series that this is the 1st session of and this is where you might be interested. This is where the recordings will be posted too.01:13:15After the fact, so you, when you go there, you'll have the chance to click on registration and register for the upcoming sessions, but also check back a little bit later, once it's gone through some processes, the recordings of this video.01:13:34And the future sessions will also be posted there at those links.01:13:50So I'm going to check to see if we do have any Q and A Now is the time. If you have those, please go into Slido, post any questions there. I'm not seeing any yet from our audience. We did get responses to our earlier polls. We certainly appreciate that.01:14:06We got information on what you feel is the approximate scale, of the edge remote sites you might be looking to upgrade on the horizon. And we also got several votes on the virtualization container stacks that might be the the backbone of your current or planned upgrades. What you will be seeing now.01:14:22Now in Slido is a survey along with several feedback questions there. If you can't take the time now unless you have questions, take a few minutes and answer some questions there. Your feedback is very valuable to us. We certainly appreciate that.01:14:42Here's a link there to the survey. If you want a QR code to it, if you want to go to the web page itself, this is where you can find it.01:15:05Now let me post some links for you in chat, there we go, so that should allow you to revisit that page for this series, view the recording.01:15:22Of this 1st session when it, when it does get posted along with checkout the Cisco community at large.01:15:41And another hot topic for us here at Cisco is Cisco IQ. I did post a link there where you can learn more about Cisco IQ by attending the upcoming series for that. Now I still don't see any questions that's being posted here.01:15:50So please take another minute to fill out the survey and provide us your valuable feedback.01:16:10We are seeing responses come in, so we haven't closed that yet. Again, if you're, if you're looking forward to the recordings, check back on that link that I put in chat.01:16:30If you have some specific, specific information you can give us today about our excellent presentation and demonstration, please jump in that survey and fill out those feedback questions. Give everybody a minute here to finish up. Got a couple of responses coming in.01:16:51And don't forget as well the information link that I put in chat surrounding Cisco IQ, there's a lot of information to learn there, and I think you'll find a lot of benefit in that.01:17:20Yeah unless we've got anything else here.01:17:40We can wrap up our session for today. Thank you again for the folks that are able to join in the feedback. Thank you for attending today's expert insight series session.01:17:47On unified Edge Foundations, this was part one. Please join us for part two coming up. And again, thank you for your time today.01:18:05