Multi-cluster gateway
1. Multi-cluster gateway
Next, let's explore multi-cluster gateway, a component used to control network ingress from outside a cluster. Multi-cluster gateway, or MCG, is a cloud-hosted controller that provisions Google Cloud HTTPs load balancers to route traffic across clusters and regions. It's the successor to multi-cluster ingress, or MCI, providing more advanced features and aligning with the Kubernetes Gateway API standard. Multi-cluster gateway ensures simple, secure, and scalable management of application networking across clusters. Examples of its routing capabilities include traffic splitting, traffic mirroring, and health-based failover. Multi-cluster gateway is based on the Kubernetes Gateway API and it's designed to be role-oriented, portable, expressive, and extensible. The Google Gateway API is the GKE gateway controller, and it's available for single or multi-clusters. This Google-hosted controller is used to manage the underlying cloud load-balancing resources. The Gateway API has several key resources, including gateway class, gateways, routes, and policies. Let's examine these resources in detail. Gateway class provides a template for creating load balancers. It defines the type of load balancer and its capabilities. In GKE, different gateway classes correspond to different Google Cloud load balancers. Gateways define where and how load balancers listen for traffic. Cluster operators create gateways based on a gateway class. HTTP routes define rules for routing HTTP(S) requests from a gateway to Kubernetes services. Policies define implementation-specific characteristics of a gateway resource-- for example, health checks and front end or back end configurations. Policies are attached to gateways to configure specific behaviors.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.