Appendix D. Cluster Federation
In the section about high availability in chapter 11 we explored how Kubernetes can deal with failures of individual machines and even failures of whole server racks or the supporting infrastructure. But what if the whole datacenter goes dark?
To make sure you’re not susceptible to datacenter-wide outages, apps should be deployed in multiple datacenters or cloud availability zones. When one of those datacenters or availability zones becomes unavailable, client requests can be routed to the apps running in the remaining healthy datacenters or zones.
While Kubernetes doesn’t require you to run the Control Plane and the nodes in the same datacenter, you’ll almost always want to do that to keep network latency between ...
Get Kubernetes in Action now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
