Nope. There is python package which deals with multi-core programming.
(I suppose through creating different processes).
Just look at http://www.parallelpython.com/

This is a really nice series, although if my experience tells me anything, seeing this kind of programming used by regular bioinformatics hackers is still some way off. Really it shouldn’t be, most high-end laptops/desktops are dual core, why not take advantage of two (four) processors ?

It is also interesting that you chose to use python, no objections from me, it is my preferred language. I have even dabbled a little with threading on my laptop to run two parallel blast jobs using python. In the end I managed to succeed (watching both cores max out on sysmonitor), but I could never quite figure out why ? (as an aside: blast is probably a more appealing example for bioinformatics).

First, my understanding of python threads is that they are not separate system level processes, but some kind of fake process that are python specific ? Trouble is that I see two separate process when I launch two Blast runs via threading ?

The other aspect of threading that I’m still not entirely clear about is how the global interpreter lock (GIL) fits into the picture. I get resource locking to prevent race conditions, but is the GIL also invoked each time an action that manipulates memory takes place in a thread ? I’ve heard this property of python makes it unsuitable for multi-core programming ?

Anyway, the article is great, and probably deserves to be published as a tutorial in its entirety. In fact PLOS (I don’t know which journal) has a tutorial section…

If you are interested you can have a look at it, in its current form at

http://cvs.biopython.org/cgi-bin/viewcvs/viewcvs.cgi/biopython/Bio/PopGen/Async/?cvsroot=biopython

This is currently working, but please note that it will still be changed and it is not a public release yet.

Other than the next blog post in the series, I plan to formally document the above code within the next month or so.