search

juji-io / datalevin

A simple, fast and versatile Datalog database
https://github.com/juji-io/datalevin
Eclipse Public License 1.0
1.07k stars 60 forks source link

Use CVM algorithm for faster cardinality estimation #254

Open huahaiy opened 1 month ago

huahaiy commented 1 month ago

As discussed in https://news.ycombinator.com/item?id=40379175

This reduces planning time for queries involving attributes with huge number of distinct values.