GKE Autopilot and GKE standard
1. GKE Autopilot and GKE standard
Now let's explore the two available modes of operation, Autopilot and Standard mode, in more detail. At a high level, Autopilot mode optimizes the management of Kubernetes with a hands-off experience. However, less management overhead means less configuration options. And with Autopilot GKE you only pay for what you use. The Standard mode allows the Kubernetes management infrastructure to be configured in many different ways. This requires more management overhead, but produces an environment for fine-grained control. With GKE standard, you pay for all of the provisioned infrastructure, regardless of how much gets used. Let's examine the benefits and functionality of Autopilot in more detail. Autopilot is optimized for production. As a Google-managed and optimized GKE instance, the job of Autopilot is to create clusters according to battle-tested and hardened best practices. Autopilot defines the underlying machine type for your cluster based on workloads, which optimizes both usage and cost for the cluster and adapts to changing workloads. And without the cluster management overhead, Autopilot lets you deploy production-ready GKE clusters faster. Autopilot also helps produce a strong security posture. Google helps secure the cluster nodes and infrastructure, and it eliminates infrastructure security management tasks. By locking down nodes, Autopilot reduces the cluster's attack surface and ongoing configuration mistakes. Autopilot promotes operational efficiency. Google monitors the entire Autopilot cluster, including control plane, worker nodes and core Kubernetes system components. This way, it ensures that Pods are always scheduled. This level of monitoring allows Google to always keep clusters up to date. Autopilot also provides a way to configure update windows for clusters to ensure minimal disruption to workloads. With Autopilot, Google is fully responsible for optimizing resource consumption. This means you only pay for Pods, not nodes. Now that you've seen many of the Autopilot mode benefits, let's look at some restrictions presented with this operation mode. The configuration options in GKE Autopilot are more restrictive than in GKE Standard. This is because GKE Autopilot is a fully managed service and has a pod-scheduling service level agreement. Autopilot clusters also have restrictions on access to node objects. Features like SSH and privilege escalation were removed and there are limitations on node affinity and host access. However, all Pods in GKE Autopilot are scheduled with a Guaranteed class Quality of Service (or QoS). But this requires a minor configuration change. The GKE Standard mode has the same functionality as Autopilot, but you're responsible for the configuration, management, and optimization of the cluster. Unless you require the specific level of configuration control offered by GKE standard, it's recommended that you use Autopilot mode.2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.