8 comments

[ 3.3 ms ] story [ 45.5 ms ] thread
"Unlike xargs however, Parallel lets me make use of the many cores that I have access to [...]"

Unlike xargs? That's just plain wrong, see xargs' -P flag.

xargs can't distribute tasks to remote CPUs though.
Oh man ! I wish someone had told me this before. Back when i did research for my BSc thesis, I had an 24 core machine and wrote a script to create 24 bash scripts to split up my task which i would than run in parallel.

Sometimes I have the felling that I am wasting time when I check HN too often, but stuff like this makes me think otherwise.

You could have also used "make -j", depending on your specific task.
This is sometimes true.

I think that Parallel is more interesting as an instant server-farm for "taco bell programmers" -- q.v. http://teddziuba.com/2010/10/taco-bell-programming.html and http://news.ycombinator.com/item?id=1818816 for the HN discussion -- as per Mr. Dziuba's definition.

Like, async spidering with parallel is a one-liner, but comparatively, have you seen how much absurdly syntaxed pycurl boilerplate you need? It's crazy. I for one will take the simpler method any day.

That is just going to thrash hell out of your disk unless it's executing on distinct machines with data well spread over a shared filesystem.