Consider using default sorting algorithm with partition.sort_values function

[sarahyurick]

Hi, I was wondering if there is a reason behind kind="mergesort" being used in sort.py:

        partition = partition.sort_values(
            by=[col], ascending=asc, na_position=na_position, kind="mergesort"
        )

instead of not specifying the kind parameter and using the default quicksort? There also does not appear to be a way to override this.

[nils-braun]

Hi @sarahyurick - thanks for your question! Yes, there was a reason behind this indeed. It is described a few lines above the code you pasted and is basically a speed improvement when we have multiple columns (referencing this pandas issue).
Do you see any performance degrading? I would be very interested to learn about that!

[sarahyurick]

Thanks for the response! I was mostly just curious - I haven't compared Pandas' mergesort versus quicksort performance before.