Hi, everyone. My name is Thomas Johnson with the customer experience group here at Cisco. Today, I'm going to show you a fundamentally different way to design and operate your data center network.00:06And then let you try it for free. We'll spend a few minutes on background, then I'll do a live demonstration of the designer tool, designing a real fabric topology and exporting a cabling plan.00:18I'll close with how switch claiming works, which is where the design becomes a real running fabric. You already know the Nexus platform. Hyperfabric is what happens when you take that hardware.00:35And pair it with a cloud operating model. You get the performance and reliability you're used to, but the deployment and operations experience is completely transformed. The designer tool is free with your Cisco cco ID.00:50No hardware required. You can design your topology or leverage a number of predefined templates. Get a complete cabling plan with specific port and transceiver assignments and hand that to your cabling team before the switches even ship.01:07Four platforms are shipping today, including the new Cisco 6100 series built specifically for hyper fabric. And three nexus 9300 models are coming in 2nd half of this year. So if you already have a 9336C.01:25Or 9031 80 FX three, or you want to start with lower cost copper access ports. That path is on the roadmap. Let's jump into a demonstration.01:43You'll want to log into hyperfabric at hyperfabric.cisco.com. If you don't have an account, you can sign up here. Otherwise log in with your email address.01:58If this is the 1st time you've logged into hyper fabric, you'll be prompted to create an organization. An organization is what stores all of your configurations and objects. You can be a member of more than one organization and they're listed over here.02:17For this demonstration, I'll be using this organization. To get started, let's create a new fabric. Click add new fabric. You can either design a fabric from scratch using the designer or you can choose from templates that have already been configured for you.02:34We're gonna do that. You have two tabs up here, one for switched fabrics and another for AI clusters. This demonstration will be using switched fabrics, and I'll use this min spine leaf template.02:521st step here shows the fabric name as well as the tiers and connections. We'll edit the fabric name to be something a little more logical This is a two tier network where you have a tier of spines in.03:14In this case there's a quantity two and a tier of leafs. Quantity four. On the left are the device details of the hardware that's been selected, as well as some configuration items like the port side, exhaust or power supplies.03:34If you scroll down, you can look at your connections. This template includes two connections between each leaf and spine, and there's NO connections between the spines themselves or between the leaves. Clicking on.03:54The connection group, you can see which pluggables were selected and how many between those devices. In this case, I'll change the pluggable to be a little bit longer for my data center.04:11Once we've got the correct pluggables.04:36We'll save the fabric blueprint. We're going to continue to the topology, but we'll come back for the cabling. We're now looking at the main fabric page. We can expand.04:38The topology to view the network, we can also edit some details of the fabric If we scroll down, we can look at the device inventory.04:54If you recall, we have two spines and four leaves. If you want to change the host names, you can click on the device and modify the host name at the top.05:13You could also modify some of the details of the device such as the rack that it's in. We'll edit 2nd leaf as well.05:30On this page, you can also review the bill of materials. This is all the equipment that's being used in this fabric.05:59You can export to a CSV for a list of all the PIDs. You can also request an estimate that can be shared with your partner for order completion. When you're ready to save the changes, you can click review configuration.06:08This is the list of configuration changes which can also be viewed line by line with the JSON diff. To push it, you need to give it a comment.06:27And then push the configuration. You can see here there's a button to switch to edit mode. Right now we're in running mode where we won't be able to make any changes, which is the default mode.06:40For users with read only access. Click here to switch to edit mode to make changes to the fabric. Let's go ahead and create the cabling plan.06:55Clicking on run cabling will generate a list of all the cables that need to be run. Down here on the cabling tab, you can see you have the connection list.07:11This was based on two connections per leaf to each spine. You can click on the exports CSV to generate a file with all the connection settings that you could then hand off to your data center installation team. On the left is where you configure.07:26Other fabric settings such as.07:45Vrfs, logical networks like interface svis, assigning host ports to VLANs, or modifying the port configuration for configuring host connections or routed connections.07:49Again, reviewing the configuration allows you to see the proposed changes. You can then push those with a comment.08:12And push the configuration. Let's talk a little bit about binding and claiming the switches. Installing the devices is optimized for quick and easy deployment.08:29No need to load code ahead of time or assign management IP addresses. Claiming a device securely associates it to your organization, and binding it assigns it the configuration you already created in the blueprint.08:40There's NO CLI, NO pre staging in a staging rack. You plug the switch in, it finds the cloud, you enter a claim code or click claim discovered devices, and within minutes, your fabric is running. The blueprint you designed.08:57All the way down to which ports connect where drives everything automatically. Here you can see how to claim a switch using the claim code, which can be found when connected to the console of the switch. You can claim all your switches using the claim code.09:13But once you've claimed one switch and have the others physically connected, you can claim the rest through the discovered devices page. Just validate the serial numbers and you are up and running. Now let's look at the claiming and binding process.09:30To bind a physical device to a logical node, you can view the topology page, click on a node, and then bind or you can go down to the device inventory, click on a node.09:48Again, click bind. You can select an available device that you'd like to bind based on the serial number. You can either use the configuration from the blueprint if you've already defined what management IP information.10:01Should be assigned or use the configuration from the device. You'll click bind, you need to review the configuration.10:20And push it. It may take a couple of minutes, but if you notice the node now has a solid line around it and if you click on the device.10:36You'll see some of the physical information start to populate, like the fan status and power supplies. You'll also see that its status is bound. Everything I just did took less than 10 min and I could do it at home tonight with NO hard.10:57Are in hand. When the switches arrive, they connect to the internet, are claimed and configured, and the fabric comes up automatically. Underlay, BGP, VXLAN, all of it. Configured by the cloud. Log into hyper fabric today to see just how easy it can be to deploy a data center network.11:16These sessions from the Cisco Live on demand library provide additional information on getting started with hyper fabric.11:36
