Single Node processing — Spark, Dask, Pandas, Modin, Koalas vol. 1

Tomas Peluritis
Uncle Data
Published in
5 min readNov 29, 2019

--

For a long time, I’ve been hearing and seeing in blog posts — “use Pandas/Spark/Dask; it’s better than the others.” From my point of view, it was precisely a stalemate, where anything could happen. Finally, I got bored of hearing the same over and over again, so I’ve decided to benchmark the most popular ones and allow myself to know in which case use what.

--

--

Tomas Peluritis
Uncle Data

Professional Data Wizard— Data Engineering/DWH/ETL/BI/Data Science.