GKE production cluster requirements

You are currently viewing version 1.1 of the Apigee hybrid documentation.This version is end of life. You should upgrade to a newer version. For more information, seeSupported versions.

This section describes the basic requirements for setting up a Google Kubernetes Engine (GKE) cluster suitable for an Apigee hybrid production installation.A production cluster can consume significant resources because it includes more nodes and larger virtual machine configurations than a non-production setup. Also, a production cluster might be hosted in multiple zones and regions worldwide.

Creating the cluster

For information about creating GKE clusters, see Creating a cluster in the GKE documentation.

Production cluster requirements

Follow these requirements when creating your production cluster:

Software versions

RequirementValue
Kubernetes versionVersion 1.13 or laterYou can find the supported Kubernetes cluster versions by issuing the following command:
gcloud container get-server-config
 Supported versions are returned under thevalidMasterVersions: heading.
kubectl versionVersion 1.13.x or later. See Install and Set Up kubectl.

Cluster template configuration

Choose these template configuration options for your production GKE cluster:

ConfigurationValue
Cluster templateStandard cluster
Location typeRegional (recommended)
Availability zones3
AutoscalingEnable autoscaling for your nodes, with the minimum and maximum number of instances set to the same number. GKE will give you a warning about setting the values to the same number. You can ignore this warning.

For more information, seeScale and autoscale hybrid services.

StorageStorage is provisioned dynamically using theStatefulSet dynamic provisioner.

Node requirements

A node pool is a group of nodes within a cluster that all have the same configuration. By default, you must configure two node pools in your cluster -- one for the stateful Cassandra pods and one for the rest of the pods, which are stateless. Configure these two node poolsas described below:

Stateful node pool configuration

Configure a stateful node pool as follows:

ConfigurationValue
Nameapigee-dataNOTE: You must use the nameapigee-data. If you want to use another node pool name, seeConfigure dedicated node pools.
Minimum number of nodes3
Machine typeN1-standard-8 (Standard machine type with 8 vCPUs, 30 GB of memory)
StorageStorage will be provisioned dynamically usingStatefulSet dynamic provisioner

Stateless node pool configuration

Configure a stateless node pool as follows:

ConfigurationValue
Nameapigee-runtimeNOTE: You must use the nameapigee-runtime. If you want to use another node pool name, seeConfigure dedicated node pools.
Minimum number of nodes3 (1 per zone)
Machine typeN1-standard-4 (Standard machine type with 4 vCPUs, 15 GB of memory)

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2026-02-19 UTC.