Skip site navigation (1) Skip section navigation (2)

Postgres vs. intel ccNUMA on Linux

From: James Robinson <jlrobins(at)socialserve(dot)com>
To: Hackers Development <pgsql-hackers(at)postgresql(dot)org>
Subject: Postgres vs. intel ccNUMA on Linux
Date: 2010-09-29 18:45:20
Message-ID: BF587806-E961-4AB8-9ED9-164F3BA55E75@socialserve.com (view raw or flat)
Thread:
Lists: pgsql-hackers
Hackers,

	Any tips / conventional wisdom regarding running postgres on large- 
ish memory ccNUMA intel machines, such as a 32G dual-quad-core,  
showing two NUMA nodes of 16G each? I expect each postgres backend's  
non-shared memory usage to remain nice and reasonably sized, hopefully  
staying within the confines of its processor's local memory region,  
but how will accesses to shared memory and / or buffer cache play out?  
Do people tune their backends via 'numactl' ?

	Furthermore, if one had more than one database being served by the  
machine, would it be advisable to do this via multiple clusters  
instead of a single cluster, tweaking the processor affinity of each  
postmaster accordingly, trying to ensure each cluster's shared memory  
segments and buffer cache pools remain local for the resulting backends?

Thanks!
----
James Robinson
Socialserve.com


Responses

pgsql-hackers by date

Next:From: Dimitri FontaineDate: 2010-09-29 19:09:05
Subject: Re: recovery.conf location
Previous:From: Colin 't HartDate: 2010-09-29 18:27:11
Subject: Re: documentation udpates to pgupgrade.html

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group