What are the benefits of kubernetes / k8s vs. a simple container for datahub? k8s looks much more scalable but when does the scaling need kick in? We are trying to deploy to 30-100 users, so probably no more than 100 simultaneous users, and realistically, probably no more than 40 at one time. I don't think we'll have metadata for more than 1000 tables * 100 columns.

This content can't be displayed.

@brainy-tent-14503 can speak to this!

DataHub is several containers, some of which are only run during installation and upgrades. Running on k8 provides a managed installation and upgrade process via the helm charts. Second, is reliability. A single node is a single point of failure. If you’re ok with some downtime and having to manually run multiple containers during upgrades, then it might not be a concern for you. So in short, ease of use from using helm to perform installation and updates and reliability in case of a single node fails come to mind.