Skip to main content
New

Ideas for improving data virtualization

Related products:FME Flow
  • November 19, 2025
  • 1 reply
  • 25 views

koenterralytics
Contributor
Forum|alt.badge.img+10

I've been working with data virtualization for a few projects now. I like the possibilities, but I think it still needs some improvements to make it easier to use the functionality:

  1. There seems to be only one way of linking a data virtualization workspace to an endpoint. And it seems to be a 1:1 relation. I think it should be much more flexible which workspaces can be linked to an endpoint. You're really bound to one way of working. It would be great if you can just create a normal workspace in Form, add an endpoint from there, and link it to the virtualization API endpoint afterwards.
  2. When testing with feature caching enabled, it asks for the input json every time again. It is not obligatory to add this input (since it uses the cache), but I do not understand why you keep getting the screen. I would expect that it works the same as a parameter. As soon as it is set (and you don’t change it) it is kept in memory without showing the pop-up every time again.
  3. If I'm correct it is not possible to change the namespace of an API after creating it. This is no issue if you just started, but for maintenance in the future this is really something which should be possible.

1 reply

redgeographics
Celebrity
Forum|alt.badge.img+62

Great suggestions ​@koenterralytics! If I may add something here:

You can assign Data Virtualization jobs to a specific queue. However, this routing rule can only be seen if you’re looking at that specific DV api, not in the Engine Management. So this can potentially be overlooked if a new admin takes over (or if there’s only infrequent management necessary)