Talk will show in details what are the steps of the most famous open source tool for Kubernetes backup, Velero. By examining the details of it, we will show some strange design choices (technical debt to be more precise) and examples how to correct them and speed up your backups. As the consequence of that, we will ensure that your master nodes are protected from the continuous load of questions “Give me all your definitions” and “Explain to me resource A..” etc. Talk will include details about Kubernetes discovery cache and what are the limits on the numbers of your custom resource definitions (CRDs).

References:
https://jonnylangefeld.com/blog/the-kubernetes-discovery-cache-blessing-and-curse
https://github.com/vmware-tanzu/velero