Re:How to speed up massive data analysis by eliminating disk...
by jbl on Friday January 01, 2010 @10:53PM
(#1026732)
You can even combine this approach with split and parallel make to do a cheap single-machine parallel sort. I use a little script that generates a Makefile that can be called with make -j n that splits an input file, sorts the parts, and then merges them with sort -m. It's proved to be quite handy.
[ Parent
| on HN ]
There are no comments on this comment yet.