Google launches Autopilot for Kubernetes

Google, one of Kubernetes container technology inventors, intends with the service to make it easier to run and manage containers on the platform.

Software containers are a rapidly growing technology, but the system is not the easiest to manage, according to Drew Bradstock, product lead for Google Kubernetes Engine (GKE).

‘Despite six years of progress, Kubernetes is still proving to be very complex,’ he wrote when announcing Autopilot. ‘And what we’ve seen over the past year is that a lot of companies do welcome Kubernetes with open arms, but then they come up against its complexity.’

Google launches Autopilot for Kubernetes.

The Kubernetes platform, which Google developed but is now open source, is the most popular of the container platforms. Apps are designed in a ‘container’ to be deployed without too many problems in various environments and clouds. However, it is a question of taking the ‘without too many issues with all the necessary caution.

There are managed and ‘managed.’
GKE is already primarily a ‘managed service’, but Google is now also releasing Autopilot, a deployment service for GKE, which automatically adds a new thin layer of management. One of the differences between the two lies, among other things, at the management level. Kubernetes works with nodes (individual servers), clusters (a series of physical or virtual servers), containers (on which programs run), and pods (a group of one or more boxes on a node). While GKE manages at the cluster level, Autopilot also includes nodes and pods in its management tool.

Google explains the whole concept in its documentation. As the name already suggests, Autopilot above all means that many parameters must be preconfigured so that administrators have less work. The best practices for running and securing these clusters and nodes are also directly integrated. Autopilot’s goal is to take care of all the day two operations’ as they are called, namely the actual functioning of the containers. This is about scalability, upgrades and maintenance, for example. For developers who are implementing new containers, there isn’t much that changes.