CPU Affinity, was RE: [Zope] Linux vs. UNIX vs. BSD

[Matthew T. Kromer]

sean.upton@uniontrib.com wrote:

>Hey Matt, while we are on the subject, I just wanted to make sure I had
>something right (and figured you might know): 
>
>I know that affinity helps a single Zope instance on an SMP box, but how
>about 2?  With Python's global interpreter lock (or whatever makes
>Python/Zope best suited to bind to a given CPU) - am I still going to run
>into problems if I am using the same Python to run 2 Zope instances in
>different software/instance homes, with all respective processes for each
>given Zope bound (via affinity) to respective CPUs?  Or is this okay?  I
>have some existing 2P AthlonMP and Xeon boxes, and I'd like to get the most
>zoom out of their horsepower as possible...
>
>Any thoughts are greatly appreciated.
>
>Thanks,
>Sean
>  
>


As long as *each* Python process (and thus each GIL) is processor bound, 
you'll be OK.  Any python that isnt pbound (on Solaris) or have its 
processor dispatch mask set (on Linux) on a multiprocessor machine will 
end up slowing down when the machine is partially idle.

A machine that's going full-throttle isn't as bad, curiously enough -- 
because the other CPU's are busy doing real work, the GIL doesn't have 
as much opportunity to get shuffled between CPUs.  On a MP box it's very 
important to set sys.setcheckinterval() up to a fairly large number, I 
recommend pystones / 50 or so.  You can make the number higher still 
without much concern.

There ought to be a practical upper limit for sys.setcheckinterval() 
(because eventually you'll enter a routine that yeilds the lock anyway) 
but I dont know what that is.  I want to say the minimum bytecode path 
thru Zope is about 24,000 bytecodes, but that's just a WAG -- but that 
would represent the "insanely high" outer bound for sys.setcheckinterval().

Why mention the check interval?  On a busy enough system, increasing 
sys.setcheckinterval() decreases the amount the GIL is released, and as 
such, helps so that you can run more than one Zope on a multiprocessor 
machine *without* processor binding.

I found, when I did tests about 2 years ago (wow, long time ago!) that I 
got diminishing returns above pystones/50, but below that I got good 
results.  You may discover that you can get performance close enough to 
processor binding that you chose not to bind.

-- 
Matt Kromer
Zope Corporation  http://www.zope.com/