Not every query needs to be live to the millisecond, and not every query can afford to be slow. SemanticFed is designed to let you make that trade-off per model instead of all-or-nothing.
For hot paths — an executive dashboard, a high-traffic embedded chart — you can enable result caching or materialization with a freshness policy you set. For everything else, queries stay live and federated against the source.
Because caching is an opt-in optimization on top of the same semantic model, you get speed where it counts without forking your definitions or standing up a separate copy of the world.
Pre-launch scenario; describes the caching model being built, not measured performance.
Do it yourself
Turn on result caching for a hot saved query with a freshness window you control, while everything else stays live and federated.
Open the Query Console and select the saved query that backs a high-traffic dashboard.
You should see: The saved query opens with its settings.
Ready to make this your story?
