Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

for deploying multi-cluster flyte, <https://docs.flyte.org/en/latest/deployment/multicluster.html#data-plane-deployment>, can I keep the original cluster as is (running both control and the data), and just add a data cluster?

and when creating the additional cluster, I should follow <https://docs.flyte.org/en/latest/deployment/aws/manual.html> (Connect to an EKS Cluster, OIDC Provider for the EKS Cluster, Create an EKS Node Group, the latter half of Create an RDS Database where I just connect new cluster to existing RDS)? I don't need to do this  `Install an Amazon Loadbalancer Ingress Controller` or to do the SSL cert stuff again right?

as long as the main cluster can talk to the other clusters then you shouldn’t need anything else.  you’re just hitting the kube api, so as long as that’s accessible.

and you can keep the original cluster running both yeah.

Can I reuse all the aws permissioning roles ?

Is there a step needed to have the clusters talk to each other? Or as long as they have the same roles + in the same vpc that it should be ok

the additional clusters don’t need to talk to each other, they just need to talk back to admin (for eventing) but this will use the same ingress that everyone else uses.

in the cluster with both the admin address for propeller might just be the k8s service dns.  but for the additional clusters, the admin address you’ll need to configure is the ingress one.

does the data plane cluster have to in the original cluster's EKS created security group? or just the general security group's are ok

I think it should use original cluster’s EKS created security group. otherwise, you will get some connection issues