Comment by thecleaner

Sure but single node performance. This makes it not very useful IMO since quite a few data science folks work with Hadoop clusters or Snowflake clusters or DataBricks where data is distributed and querying is handled by Spark executors.