High availability for Satellite reference architecture
As described in High availability (HA) overview, HA is important for all cloud-based applications. Ready to learn more specifics about HA in the Satellite reference architecture?
High availability in your Satellite location
For HA in IBM Cloud Satellite, you need to consider HA in three main areas within your Satellite location. For more information, see the following topics:
- Satellite control plane master
- Satellite control plane worker nodes
- IBM Cloud services that run in your Satellite location such as Red Hat OpenShift on IBM Cloud
In addition, you own all of the underlying infrastructure that the Satellite location components run on. Consider the following options to increase resiliency of that infrastructure:
- Add more compute hosts
- Add redundant power, network, and storage
- Isolate machines within one data center
- Spread hosts across physical locations
Your workloads in Red Hat OpenShift on IBM Cloud
After you have your underlying infrastructure in place, begin deploying your workloads in your on-premises Red Hat OpenShift on IBM Cloud workload clusters. You can use the cluster-autoscaler add-on to scale the worker pools in
your cluster automatically to increase or decrease the number of worker nodes in the worker pool based on the sizing needs of your scheduled workloads. The cluster-autoscaler add-on is based on the Kubernetes Cluster-Autoscaler project.
For more information, see Autoscaling clusters.
In addition, it is possible to autoscale your pods. For more information, see Scaling apps.
HA for IBM Cloud services
Your high availability strategy needs to consider all of the IBM Cloud services in your deployment, including the services that run in IBM Cloud outside of your Satellite location. For more information, see High availability for IBM Cloud services.