LATEST VERSION: v1.3 - RELEASE NOTES
Pivotal Container Service v1.1

GCP Prerequisites and Resource Requirements

Page last updated:

This topic describes the prerequisites and resource requirements for installing Pivotal Container Service (PKS) on Google Cloud Platform (GCP).

Prerequisites

Before you install PKS, you must install one of the following:

  • Ops Manager v2.1.0 or later
  • Ops Manager v2.2.0 or later

Note: You use Ops Manager to install and configure PKS. Each version of Ops Manager supports multiple versions of PKS. To confirm that your Ops Manager version supports the version of PKS that you install, see PKS Release Notes.

You must also create service accounts for Kubernetes master and worker nodes and create a load balancer to access the PKS API.

Install and Configure Ops Manager

To install an Ops Manager version that is compatible with the PKS version you intend to use, follow the instructions in the corresponding version of the Ops Manager documentation.

Version
Ops Manager v2.1
Ops Manager v2.2

Create Service Accounts for Kubernetes

After you install and configure Ops Manager, you must create service accounts for Kubernetes master and worker node VMs in your PKS deployment. To create the service accounts, follow the procedures in Creating Service Accounts in GCP for PKS.

Create a Load Balancer for the PKS API

Before you install PKS, you must create an external TCP load balancer so that you can access the PKS API from outside the network. This load balancer allows you to run pks commands from your local workstation. You must create the load balancer before you install PKS, and then complete the load balancer configuration after you install PKS.

To create a load balancer in GCP, follow the procedures in Creating a GCP Load Balancer for the PKS API.

Resource Requirements

Installing Ops Manager and PKS requires the following virtual machines (VMs):

VM CPU RAM Storage
Pivotal Container Service 2 8 GB 16 GB
Pivotal Ops Manager 1 8 GB 160 GB
BOSH Director 2 8 GB 16 GB

Each Kubernetes cluster provisioned through PKS deploys the VMs listed below. If you deploy more than one Kubernetes cluster, you must scale your allocated resources appropriately.

VM Name Number CPU Cores RAM Ephemeral Disk Persistent Disk
master 1 2 4 GB 32 GB 5 GB
worker 1 2 4 GB 32 GB 50 GB

Please send any feedback you have to pks-feedback@pivotal.io.

Create a pull request or raise an issue on the source for this page in GitHub