The production topology uses a leaf-spine fabric for performance and scalability. SmartFabric Services (SFS) automates the deployment of this fabric.
With SFS, two leaf switches are used in each rack for redundancy and performance. A Virtual-Link Trunking interconnect (VLTi) connects each pair of leaf switches. Every leaf switch has an L3 uplink to every spine switch. Equal-cost multi-path routing (ECMP) is leveraged to use all available bandwidth on the leaf-spine connections.
SFS uses BGP-EVPN to stretch L2 networks across the L3 leaf-spine fabric. This configuration allows for the scalability of L3 networks with the VM mobility benefits of an L2 network. For example, a VM can be migrated from one rack to another without the need to change its IP address and gateway information.
The example in this guide builds the SmartFabric shown in Figure 7 in two stages:
- The first stage is a single rack deployment. Leaf switches 1A and 1B are deployed in Rack 1 without spine switches, and a two-leaf fabric is created using SFS. The fabric is connected to the external network using either L2 or L3 uplinks. The external network is typically a preexisting network in the data center. Three VxRail nodes are connected to the two leaf switches, and a three-node VxRail cluster is deployed.
- In the second stage, two spine switches are added and connected to leaf switches 1A and 1B. Leaf switches 2A and 2B are added in Rack 2 and are also connected to the spine switches. The fabric is expanded to include the two spines and two additional leafs using SFS. A fourth VxRail node is added in Rack 2 and joined to the existing VxRail cluster.