Watch on-demand the newest GKE innovations and exclusive technical demos

Jump to

Google Kubernetes Engine

Secured and fully managed Kubernetes service with revolutionary autopilot mode of operation.

New customers get $300 in free credits to spend on Google Cloud during the first 90 days. All customers get one autopilot or zonal cluster per month for free, not charged against your credits.

  • action/check_circle_24px Created with Sketch.

    Start quickly with single-click clusters and scale up to 15000 nodes

  • action/check_circle_24px Created with Sketch.

    Leverage a high-availability control plane including multi-zonal and regional clusters

  • action/check_circle_24px Created with Sketch.

    Eliminate operational overhead with industry-first four-way auto scaling and release channels

  • action/check_circle_24px Created with Sketch.

    Secure by default, including vulnerability scanning of container images and data encryption

  • action/check_circle_24px Created with Sketch.

    Integrated Cloud Monitoring with infrastructure, application, and Kubernetes-specific views


Speed up app development without sacrificing security

Develop a wide variety of apps with support for stateful, serverless, and application accelerators. Use Kubernetes-native CI/CD tooling to secure and speed up each stage of the build-and-deploy life cycle.

Streamline operations with release channels

Choose the channel that fits your business needs. Rapid, regular, and stable release channels have different cadences of node upgrades and offer support levels aligned with the channel nature.

Manage infrastructure with Google SREs

Get back time to focus on your applications with help from Google Site Reliability Engineers (SREs). Our SREs constantly monitor your cluster and its computing, networking, and storage resources.

Key features

Key features

Autopilot mode of operation

Optimized cluster with pre-configured workload settings offering a nodeless experience. Let Google take care of the underlying infrastructure of your entire cluster, including nodes. Maximize operational efficiency and bolster security of your applications by restricting access only to Kubernetes API and safeguarding against node mutation. Pay only for your running pods, not system components, operating system overhead or unallocated capacity.

Pod and cluster autoscaling

Horizontal pod autoscaling based on CPU utilization or custom metrics, cluster autoscaling that works on a per-node-pool basis and vertical pod autoscaling that continuously analyzes the CPU and memory usage of pods and dynamically adjusts their CPU and memory requests in response. Automatically scales the node pool and clusters across multiple node pools, based on changing workload requirements.

Kubernetes applications

Enterprise-ready containerized solutions with prebuilt deployment templates, featuring portability, simplified licensing, and consolidated billing. These are not just container images, but open source, Google-built, and commercial applications that increase developer productivity, available now on Google Cloud Marketplace.

Workload and network security

GKE Sandbox provides a second layer of defense between containerized workloads on GKE for enhanced workload security. GKE clusters natively support Kubernetes Network Policy to restrict traffic with pod-level firewall rules. Private clusters in GKE can be restricted to a private endpoint or a public endpoint that only certain address ranges can access.

View all features


What’s new



Best Practice
Best practices for operating containers

Learn best practices for operating containers in GKE.

GKE tutorial: Deploying a containerized web application

This tutorial shows you how to package a web application in a Docker container image and run that container image on a GKE cluster.

Cloud Architect learning path

Learn the skills needed to modernize existing applications running on virtual machines while deploying cloud-native apps on containers.

Best Practice
Preparing a GKE environment for production

Follow the guidance and methodology for onboarding your workloads more securely, reliably, and cost-effectively to GKE.

Hardening your GKE cluster

Learn how to implement the guidance for hardening your GKE cluster.

GKE resources

Find more information on features, updates, pricing, and more.

Explore what you can build on Google Cloud

Discover Google Cloud technical resource guides to help unlock the potential of GKE.

Use cases

Use cases

Use case
Continuous delivery pipeline

Enable rapid application development and iteration by making it easy to deploy, update, and manage your applications and services. Configure GKE, Cloud Source Repositories, Cloud Build, and Spinnaker for Google Cloud services to automatically build, test, and deploy an app. When the app code is modified, the changes trigger the continuous delivery pipeline to automatically rebuild, retest, and redeploy the new version.

Diagram showcasing how developers can build a continuous delivery pipeline
Use case
Migrating a two-tier application to GKE

Use Migrate for Anthos to move and convert workloads directly into containers in GKE. Migrate a two-tiered LAMP stack application, with both application and database VMs, from VMware to GKE. Improve security by making the database accessible from the application container only and not from outside the cluster. Replace SSH access with authenticated shell access through kubectl. See container system logs through automatic Cloud Logging integration.

Diagram showcasing how to move and convert workloads into GKE

All features

All features

Identity and access management Control access in the cluster with your Google accounts and role permissions.
Hybrid networking Reserve an IP address range for your cluster, allowing your cluster IPs to coexist with private network IPs via Google Cloud VPN.
Security and compliance GKE is backed by a Google security team of over 750 experts and is both HIPAA and PCI DSS compliant.
Integrated logging and monitoring Enable Cloud Logging and Cloud Monitoring with simple checkbox configurations, making it easy to gain insight into how your application is running.
Cluster options Choose clusters tailored to the availability, version stability, isolation, and pod traffic requirements of your workloads.
Auto scale Automatically scale your application deployment up and down based on resource utilization (CPU, memory).
Auto upgrade Automatically keep your cluster up to date with the latest release version of Kubernetes. Kubernetes release updates are quickly made available within GKE.
Auto repair When auto repair is enabled, if a node fails a health check, GKE initiates a repair process for that node.
Resource limits Kubernetes allows you to specify how much CPU and memory (RAM) each container needs, which is used to better organize workloads within your cluster.
Container isolation Use GKE Sandbox for a second layer of defense between containerized workloads on GKE for enhanced workload security.
Stateful application support GKE isn't just for 12-factor apps. You can attach persistent storage to containers, and even host complete databases.
Docker image support GKE supports the common Docker container format.
Fully managed GKE clusters are fully managed by Google Site Reliability Engineers (SREs), ensuring your cluster is available and up-to-date.
OS built for containers GKE runs on Container-Optimized OS, a hardened OS built and managed by Google.
Private container registry Integrating with Google Container Registry makes it easy to store and access your private Docker images.
Fast consistent builds Use Cloud Build to reliably deploy your containers on GKE without needing to setup authentication.
Workload portability, on-premises and cloud GKE runs Certified Kubernetes, enabling workload portability to other Kubernetes platforms across clouds and on-premises.
GPU and TPU support GKE supports GPUs and TPUs and makes it easy to run ML, GPGPU, HPC, and other workloads that benefit from specialized hardware accelerators.
Built-in dashboard Cloud Console offers useful dashboards for your project's clusters and their resources. You can use these dashboards to view, inspect, manage, and delete resources in your clusters.
Preemptible VMs Low-cost, short-term instances designed to run batch jobs and fault-tolerant workloads. Preemptible VMs provide significant savings of up to 80% while still getting the same performance and capabilities as regular VMs.
Persistent disks support Durable, high-performance block storage for container instances. Data is stored redundantly for integrity, flexibility to resize storage without interruption, and automatic encryption. You can create persistent disks in HDD or SSD formats. You can also take snapshots of your persistent disk and create new persistent disks from that snapshot.
Local SSD support GKE offers always-encrypted local solid-state drive (SSD) block storage. Local SSDs are physically attached to the server that hosts the virtual machine instance for very high input/output operations per second (IOPS) and very low latency compared to persistent disks.
Global load balancing Global load-balancing technology helps you distribute incoming requests across pools of instances across multiple regions, so you can achieve maximum performance, throughput, and availability at low cost.
Linux and Windows support Fully supported for both Linux and Windows workloads, GKE can run both Windows Server and Linux nodes.
Hybrid and multi-cloud support Take advantage of Kubernetes and cloud technology in your own data center. Get the GKE experience with quick, managed, and simple installs as well as upgrades validated by Google through Anthos GKE.
Serverless containers Run stateless serverless containers abstracting away all infrastructure management and automatically scale them with Cloud Run.
Usage metering Fine-grained visibility to your Kubernetes clusters. See your GKE clusters' resource usage broken down by namespaces and labels, and attribute it to meaningful entities.
Release channels Release channels provide more control over which automatic updates a given cluster receives, based on the stability requirements of the cluster and its workloads. You can choose rapid, regular, or stable. Each has a different release cadence and targets different types of workloads.
Software supply chain security Verify, enforce, and improve security of infrastructure components and packages used for container images with Container Analysis.
Per-second billing Google bills in second-level increments. You pay only for the compute time that you use.



One autopilot cluster or zonal cluster per billing account is free.

Cluster management fee of $0.10 per cluster/hour apply, except for Anthos clusters. User pods in autopilot clusters are billed per second for CPU cores, memory, and ephemeral storage, until a pod is deleted. Worker nodes in standard clusters accrue compute costs, until a cluster is deleted.