Page last updated:

NSX-T Data Center Edge Type and Load Balancer capacity planning

It is recommended that you have a minimum of four large Edge VMs in a configuration of two active-standby pairs (at least two instances are necessary for HA) which should be considered as the minimum for load balancing. A production grade deployment should only use large load balancers.

The type of Edge deployment is up to you based upon your application needs and system capacity. Edges as VMs offers maximum flexibility but uses resources of the cluster. Bare metal Edges perform much better than Edges in VMs. Since TA4A is a “hyperconverged” or “collapsed” model, the approach of using VMs for Edges is a easy place to start. Though VMware recommends starting with Edges as VMs, you may discover that, at some point, edge capacity will become saturated. Then you will want to add or replace them with bare-metal Edges for more capacity as necessary.

Introduced in NSX-T 2.5, users can assign edges to defined fault domains. When TKGI provisions load balancing T1 routers, the active and passive edges are assigned on separate zones automatically, which increases the resiliency.

The size of the load balancer determines the number of Virtual Servers, Pools, and Pool Members per LB instance. Learn more about configuration maximums at VMware’s Configuration Maximums site

Important Note: Since TAS for VMs and TKGI use active/standby Edge Node Load Balancer Services, you will need to consider capacity per Edge pair to derive the available number of Load Balancer Services you will need. Put another way, for each pair of Edge servers, the Load Balancer Service capacity is equivalent to just one edge instance since the other is treated as stand-by.

The number of available Load Balancer instances is directly proportional (1:1) to the maximum number of Kubernetes clusters that can be supported. Each Kubernetes cluster would use a minimum of one load balancer from an active Edge instance. Based on the type of load balancer and the Edge, the load balancer instances are fixed. As such, the resulting number of Kubernetes clusters created on the edge are also constrained by the number of free loadbalancer services on the active Edge.

Note: Maximum number of Edge nodes per Edge Cluster is 10 (as of NSX-T Data Center 2.5). If the required number of Edge nodes is greater than 10, you will have to create additional Edge clusters. If the Load Balancer service capacity is fully utilized on a given Edge pair, then install and bring up additional Edge VM instance pairs in the same Edge Cluster to handle requirements for additional load balancers (for existing or new TKGI clusters).

Learn more about configuration maximums at VMware’s Configuration Maximums site

TKGI NSX-T Container Plugin for ingress traffic

There is no native way to reach the apps you will run in TKGI unless you deploy one of the following. What you select influences load balancing choices. Kubernetes has three ways for external users to access services running in a Kubernetes cluster:

The NSX-T Load Balancer supports both layer 4 and layer 7 ingress traffic, as well as SSL encryption.

The TKGI NSX-T container plugin (TKGI NCP) integrates with NSX-T Load Balancer. The Kubernetes operator can interface with Kubernetes resources (ingress controller and load balancer) to provision and configure an NSX-T Load Balancer.

The Benefits of TKGI NCP + NSX-T integration:

  • Kubernetes users only need to manage Kubernetes resources rather than lower level network infrastructure
  • NSX-T Load Balancer provisioning and configuration automatically occurs and includes the following:
    • Performs IP assignment (from pre-assigned IP Pool)
    • Configures the routing rules between the Load balancer and the Kubernetes pods
    • Configures SSL encryption/decryption

You have a choice selecting a layer 4 or layer 7 approach based upon the type of application traffic you have. A layer 4 approach is a good choice for TCP-bound traffic types. A layer 7 approach is best for HTTP-bound traffic types. Each will be described here.

Ingress Controller and Load Balancing for TKGI

Layer 7 Approach Using Ingress Controller

Ingress Controller View a larger version of this diagram.

Domain info for ingress is defined in the manifest of the Kubernetes deployment. Here is an example.

apiVersion: extensions/v1
kind: Ingress
  name: music-ingress
  namespace: music1
  - host:
      - path: /.*
          serviceName: music-service
          servicePort: 8080

Layer 4 Approach Using LoadBalancer Service

Load Balancing View a larger version of this diagram.

When pushing a Kubernetes service with type set to LoadBalancer, NSX-T Data Center automatically creates a new virtual IP (VIP) for the service on the existing load balancer for that namespace. You will need to specify a listening and translation port in the service, along with a name for tagging. You will also specify a protocol to use. Here is an example.

apiVersion: v1
kind: Service
  type: LoadBalancer
  - port: 80
    targetPort: 8080
    protocol: TCP
    name: web