Open stroomdev66 opened 3 years ago
Consider using user-defined node attributes to be entered as a processor filter condition. A node to its attributes should be a one-to-many relationship.
Another idea is to allow the user to define a weight for a processor filter. This would add nuance to the existing priority
setting and allow users to further prioritise or restrict processor time to streams.
Also see gh-1622, gh-2330
It should be possible to choose which nodes process which data by making processor filters target node groups (gh-1623).