What is Google Container Engine?
Overview
Welcome to Google Container Engine. Inspired by Google’s experience with building and running container-based distributed systems, Container Engine re-imagines some of Google’s most powerful internal systems, so that you can develop and manage containers the way Google’s engineers do.
With container-based computing, application developers can focus on their application code, instead of on deployments and integration into hosting environments. At the same time, applications can be built with few constraints. Operations can provide a robust platform that quickly provisions compute resources and easily manages applications. The tools need to support the right controls for such application and resource management.
Our focus with Container Engine is on building these tools and controls for operations. At the same time, we want to allow for workload mobility, where containerized applications can run multi-cloud. We have, therefore, designed Container Engine to support Kubernetes, the open source technology, so that customers can run on multiple clouds.
Features and roadmap
In this first release of Container Engine, we introduce our cluster concept. A cluster is a bundle of Compute Engine resources that can be centrally managed. A cluster has its own lifecycle and dedicated resources, all backed by virtual machines.
Clusters
A cluster is comprised of a master and worker nodes. Creating a cluster results in a fully provisioned set of nodes and a master, all on a configured IP private network. The master becomes the entry point for creating and controlling the Compute resources, reached through the cluster’s API endpoint.
Nodes
Each node in the cluster provides processor and memory resources. A node belongs to only one cluster and gets provisioned and turned on during cluster creation. The number of nodes created should be based on the total amount of Compute Engine resources expected. The cluster master schedules work on each node.
In our alpha version, the number of nodes in a cluster is fixed for the lifetime of the cluster, and the maximum number of nodes is 50. All nodes in the cluster must be of the same VM instance type.
Kubernetes and the API Endpoint
The full open source Kubernetes suite runs on Container Engine. The supported versions of Kubernetes are updated frequently; refer to the release notes for the currently supported version.
Community and feedback
This alpha release is fully open to the public.
If you'd like to be involved in the design of Google Container Engine and the related open source projects:
- Give early feedback and talk with the community (including the developer team) on the google-containers mailing list.
- Ask development and best practices questions on Stack Overflow using the google-container-engine tag.
- Browse and contribute to the Kubernetes project on GitHub.
- Chat with the community on IRC.
Pricing
Container Engine uses Google Compute Engine instances for nodes in the cluster, including the cluster master (created automatically for every cluster) and the number of worker nodes you specify. You will be billed for those instances according to Compute Engine's pricing, until the clusters are deleted. During this alpha phase, there is no additional charge for Container Engine on top of the Compute Engine pricing.