How is KubeStellar different than OCM? #510
clubanderson
started this conversation in
General
Replies: 1 comment
-
@MikeSpreitzer @pdettori @ezrasilvera can you help us formulate this PoV so that we can include it in our docs |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
When compared with OCM, KubeStellar takes a slightly different approach and adds some interesting capabilities. First, we do not use a manifest to bundle resources that you intend to have delivered to your multicluster/multicloud/edge environment. OCM uses manifestwork, we allow unbundled usage of any Kubernetes resource - namespaced or clusterscoped. Second, we have the notion of upsyncing resources. This comes in handy when you have a spoke/edge that needs to ‘communicate’ back to the hub in some way other than just a status update in the original CR. Third, we have the capability to use clusters as transport rather than target. We denature resources so that they are not unpacked/applied to clusters where they are not intended to execute. This allows us to go from a 2-tier hub-and-spoke model/environmnet, to a n-tier hub-intermediary-spoke environment. Then, add the ability to customize workloads in groups and summarize statuses to reduce the cognitive load of your devops. This is really useful when you want to reach far edge clusters that are not directly connected to your hub. I think this list enumerates the differences between KubeStellar and OCM
Beta Was this translation helpful? Give feedback.
All reactions