In this model, a single vSphere cluster hosts all the NSX Edge devices while also hosting tenant workloads. When you plan to expand the physical infrastructure that is available for workloads, you use the existing initial shared edge and compute cluster to provide routing services for all new workloads.

As you deploy building blocks, you either reuse existing NSX Edge devices or deploy NSX Edge devices in the original shared edge and compute cluster to handle the north-south traffic into and out of the SDN for workloads in the new building block.

Figure 1. Single Shared Edge Cluster Model


Considerations

  • Shared edge and compute cluster is the initial configuration that Cloud Builder deploys for VMware Validated Design.

  • This model works well in environments where transports zones will span several clusters, allowing workloads to move horizontally between compute clusters without the need to change their IP addresses.

  • When scaling up the environment, examine the resource usage of the shared edge and compute cluster. Determine whether the cluster can continue to handle the north-south traffic for all workloads in the environment, and whether the workloads and edge gateways on that cluster will experience contention for compute or network bandwidth resource as the number of edges and volume of traffic increases.

  • In an NSX for vSphere environment, when deploying additional workload domains, this model is more cost effective than a dedicated edge cluster model because you deploy only a single new cluster for each new workload domain while in a dedicated edge cluster model you deploy a minimum of two clusters. For information on the dedicated edge cluster model, see Dedicated Edge Cluster.