Hacker News new | ask | show | jobs
by floki999 2418 days ago
Dask anyone? Allowed me to handle a 60 Gb dataset on a machine with 16 Gb RAM. High level of compatibility with Pandas dataframes.
2 comments
roaur 2417 days ago
I'm a Dask evangelist. It's a remarkable tool and is one of the first I reach for when this problem arises. Maybe it's not well known?
link
xiaodai 2417 days ago
I mean, don't reach for Spark. Just get Dask, a databased or R's disk.frame is mostly fine.
link