Hello 👋 I have a question regarding sampling facets (doc). We have a collection with ~2.5M documents, where we allow filtering by ~15 different fields. Given the size of the collection, we’ve enabled facet sampling by setting

facet_sample_percent

to a reasonable value (30-50). While performance in general improved and we’re happy with the tradeoff of sampled vs exact counts. But sometimes, if a facet is not found in the sample, the value won’t be available to filter on. I was wondering what you recommend for such use-cases.

We fixed a few facet performance issues in the most recent RC build. Could you try the same queries on v28.0.rc27 or above?

Hi Jason, thank you for getting back to me. I’ll give it a go and report back