[prev in list] [next in list] [prev in thread] [next in thread] 

List:       uwsgi
Subject:    [uWSGI] Request distribution among workers
From:       roberto () unbit ! it (Roberto De Ioris)
Date:       2011-07-12 6:10:36
Message-ID: eebc331f0d3dd6e1466b5a6d4bd1d9d6.squirrel () manage ! unbit ! it
[Download RAW message or body]


>>
>
> But isn't this transparent to the Python process? The goal here is to
> force
> python threads to be pigeon holed into a specific processor. I'm no
> expert,
> but I would assume that the virtual architecture would handle this
> correctly. Perhaps I assume too much!
>


Oh sorry, forgot to answer to this part, yes, setting cpu-affinity to 1
will ensure you that each cpu can run a thread concurrently reducing (a
lot) the GIL impact.


-- 
Roberto De Ioris
http://unbit.it

[prev in list] [next in list] [prev in thread] [next in thread]