We currently have a mix of Domo datasets plus federated datasets in our instance. When new cards are created, Domo often picks columns that are pretty irrelevant most of the time. While this isn't much of a big deal for Domo datasets since there is a speed advantage there, it creates a negative user experience for federated datasets. Because Domo picks what it thinks is useful, it really kills performance when federating because Domo may decide to select columns that aren't part of the source systems partition/indexing strategy. This leads to Analyzer taking a very long time to see a first result as part of new card creation.
Instead, it would be nice to have two options.
First, an admin setting that lets us turn of the analyzers auto preview system wide.
Second, have an option added to all federated/cloud amplified connectors so that if I disable auto-preview in my SQL Server, Dremio or Snokeflow federated connectors any datasets that came from those connectors won't trigger auto-preview when cards are built on top.
This would keep Domo from hammering the federated sources with bad column selections while at the same time giving flexibility to allow auto-preview from other sources (like standard Domo datasets)
Thanks!