Managing multiple Kubernetes clusters introduces challenges such as multi-cluster management, unified ingress, and cross-cluster resource scheduling. The fleet management feature of ACK One provides ACK-managed fleet instances. Use these instances to manage Kubernetes clusters in any environment for a consistent cloud-native application management experience.
Fleet console
Core concepts
A fleet instance is built on mature open source projects such as Argo CD and Open Cluster Management to ensure openness. A fleet provides developers with GitOps capabilities and a consistent Kubernetes API to schedule and distribute workloads, applications, and configurations. It also supports traffic management.
An Associated Clusters: You can add ACK clusters from the same account to a fleet as associated clusters. You can then use the fleet instance to distribute applications and workloads to these associated clusters. A fleet can manage associated clusters across regions.
A fleet instance administrator: The user who creates the fleet instance. This administrator can create namespaces, configure job priorities, and set resource quotas for other developers.
A developer: Developers can only submit jobs or publish applications in specifically authorized namespaces.
Use cases
ACK One fleet management is suitable for the following scenarios:
-
The control plane of an ACK One Fleet instance is built on the Kubernetes control plane and provides standard Kubernetes APIs.
-
You can add multiple Kubernetes clusters to an ACK One Fleet instance. If a cluster is not an ACK cluster, you must first register it.
-
GitOps for application distribution: Host open-source Argo CD to implement GitOps-based application distribution across multiple clusters.
-
Traffic management: Integrate an MSE cloud-native gateway to control north-south traffic for multi-cluster applications. Multi-cluster Services are also supported to control east-west traffic.
-
Job distribution: Schedule offline jobs to run on the most suitable cluster in a multi-cluster environment.
-
Global monitoring: Use a central dashboard to view monitoring metrics from all your clusters.