This section lists the form factors that the NSX Application Platform supports, along with the minimum resources required for each. Additional system requirements prior to deployment are also provided.

System Requirements

In addition to the information listed in NSX Application Platform Deployment Prerequisites, use the following table as a guide when working with your infrastructure administrator to prepare for deploying the NSX Application Platform and the NSX features hosted on the platform. The form factor you select determines which NSX features you can activate or install on the platform.

Before you deploy the NSX Application Platform, determine the size of the VMware TKG Cluster on Supervisor or Upstream Kubernetes cluster and the minimum number of nodes that your infrastructure administrator must allocate.

Table 1: Minimum System Requirements
Form Factor

Recommended # of Nodes in a TKG Cluster on Supervisor or Upstream Kubernetes Cluster

vCPU

Memory

Storage

Ephemeral Storage

Supported NSX Features

Advanced

  • 2 vCPUs for the control plane node (See more information later in this topic about using guaranteed-small VM)
  • 16 vCPU per worker node
  • 4 GB RAM for the control plane node (See more information later in this topic about using guaranteed-small VM)
  • 64 GB RAM per worker node

1 TB per NSX Application Platform instance

(See more information later in this topic.)

64 GB
  • Security Intelligence

  • NSX Network Detection and Response

  • NSX Malware Prevention

  • NSX Metrics

(See the Important Note later in this topic.)

Evaluation

1 control plane node and 1 worker node

  • 2 vCPUs for the control plane node (See more information later in this topic about using guaranteed-small VM)
  • 16 vCPU per worker node
  • 4 GB RAM for the control plane node (See more information later in this topic about using guaranteed-small VM)
  • 64 GB RAM per worker node

1 TB per NSX Application Platform instance

64 GB
  • Security Intelligence

  • NSX Network Detection and Response

  • NSX Malware Prevention

  • NSX Metrics

VMware Harbor OVA Requirements

VMware Harbor version 2.8.x or later is recommended. A packaged OVA is available from the Broadcom support portal under the Tanzu Kubernetes Grid section.

Enter your Broadcom support portal login credentials and click the Download product bits from the support portal icon. (HTTPS Download) icon to access the Harbor OVA file.

The Harbor OVA has the following resource requirements:

vCPU Memory Storage Networking
4 8GB

20GB for the base VMware Harbor appliance

150GB for NSX Application Platform components

1 vNIC/IP address with connectivity to NSX Manager and NSX Application Platform nodes

An existing Harbor instance or standards-based compatible repository can also be used. A minimum OCI support must be available in the repository. You must implement the specific processes for uploading and sharing Helm charts and Docker images using OCI.

Support for Minimum Four to Five Worker Nodes

Important:
When using four Advanced Form Factor worker nodes in a TKG Cluster on Supervisor or Upstream Kubernetes Cluster, only the following NSX features that are hosted on the NSX Application Platform can be activated.
  • Security Intelligence
  • NSX Metrics for NSX Application Platform (activated by default)
To activate the following NSX features, in addition to Security Intelligence and NSX Metrics for NSX Application Platform, five worker nodes in the guest cluster are required.
  • NSX Network Detection and Response
  • NSX Suspicious Traffic for Network Traffic Analysis
  • NSX Malware Prevention

All the supported features listed in System Requirements for the Advanced Form Factor size can be activated on a four to five worker node guest cluster.

Evaluation Form Factor

The Evaluation form factor is applicable only in non-production deployments for use in evaluations or demonstrations. It has limited data retention, no scale out or high availability support, and no support for upgrades. If you want to run multiple services on the Evaluation form factor for an extended period of time, increase the worker node resources to 24 vCPU, 128 GB of RAM, and 128 GB for the image file system partition used by containerd when using a single node. If necessary, contact support for more sizing details.