Real-time replication of Kubernetes cluster

General Discussions
1

Hello everyone
I’m still a Kubernetes newbie but I am already confronted with what I think is a mammoth task. My company is hosting a Kubernetes cluster. Because of internal policies, we are obliged to have everything georedundand. Which means we are supposed to build a second cluster (identical to the already existing one) in a different data center. Only ever one of them is active and the remaining one acts as a backup.
My task is to come up with different approaches on how we can synchronize those two clusters in real time, so that in case of failover operations can continue without much interruption.

From what I’ve heard such a replication can happen on different levels (hypervisor, storage, application…). I want to find out more about these different approaches but I lack the proper lingo to find any good online sources for my problem.
Do you guys know any sources who cover my question? Is there any literature on this? Or any personal experience you might want to share? Thanks in advance!

2

Are these 2 data centers location in same region?(probably few kms/miles away)
Do they have high speed redundant private connectivity between them?(latency in single digit, <10ms)

If these 2 requirements are satisfied, you can use the same cluster with worker nodes distributed across multiple data centers. Active/Backup on k8s cluster isn’t a resource friendly solution, instead take advantage of available resources on both datacentres & based on need it can be scaled up/down further.