I have some slow running models and have found adding `persist_for` for 1440 minutes and scheduling the dashboard to refresh overnight everyday to be helpful. However, when users try to filter the dashboard, I find we still face very long load times. Are there other recommendations?
Have you tested "suggest_persist_for" for the filters used in the dashboard? This would help to cache the suggestions for filters based on the: https://cloud.google.com/looker/docs/reference/param-field-suggest-persist-for
Anahita's suggestion is a good one, but if you are still looking at slow-loading content, the ultimate problem is that the queries are taking too long on the database end. The solution to this, rather than refreshing the dashboard, would probably involve pre-aggregating some data in PDTs and rebuilding your data models so that they are querying these much smaller tables and returning much faster results.