Polars
@pola.rs
Dataframes powered by a multithreaded, vectorized query engine, written in Rust.
In that blog they explain they did manual join order optimization for Dask only.
We don't allow that as that can differ hugely in rows produced by joins and would be an unfair comparison.
We don't allow that as that can differ hugely in rows produced by joins and would be an unfair comparison.
June 19, 2025 at 6:17 AM
In that blog they explain they did manual join order optimization for Dask only.
We don't allow that as that can differ hugely in rows produced by joins and would be an unfair comparison.
We don't allow that as that can differ hugely in rows produced by joins and would be an unfair comparison.
We enabled pyarrow datatypes. There is no GIL contention on them. If you think we should adapt the settings, please let us know, we want to run with optimal settings.
June 18, 2025 at 2:32 PM
We enabled pyarrow datatypes. There is no GIL contention on them. If you think we should adapt the settings, please let us know, we want to run with optimal settings.