Received: from maia.hub.org (maia-5.hub.org [200.46.204.29]) by mail.postgresql.org (Postfix) with ESMTP id BED131337B59 for ; Tue, 12 Oct 2010 11:20:28 -0300 (ADT) Received: from mail.postgresql.org ([200.46.204.86]) by maia.hub.org (mx1.hub.org [200.46.204.29]) (amavisd-maia, port 10024) with ESMTP id 97750-04 for ; Tue, 12 Oct 2010 14:20:22 +0000 (UTC) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from smtp02.lnh.mail.rcn.net (smtp02.lnh.mail.rcn.net [207.172.157.102]) by mail.postgresql.org (Postfix) with ESMTP id D79E41337B55 for ; Tue, 12 Oct 2010 11:20:19 -0300 (ADT) Received: from ms14.lnh.mail.rcn.net ([208.58.56.82]) by smtp02.lnh.mail.rcn.net with ESMTP; 12 Oct 2010 10:20:20 -0400 Received: (from ms14.lnh.mail.rcn.net [69.182.178.231]) by ms14.lnh.mail.rcn.net (MOS 4.1.9-GA) with HTTP/1.1 id ANC95779 (AUTH gnuoytr@rcn.com); Tue, 12 Oct 2010 10:20:19 -0400 (EDT) From: Subject: Re: How does PG know if data is in memory? To: pgsql-performance@postgresql.org X-Mailer: Mirapoint Webmail Direct 4.1.9-GA MIME-Version: 1.0 In-Reply-To: References: <4CA9488C.2080502@2ndquadrant.com> <4CAA5974.2050805@wizmail.org> <20101011231134.ANB76937@ms14.lnh.mail.rcn.net> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Message-Id: <20101012102019.ANC95779@ms14.lnh.mail.rcn.net> Date: Tue, 12 Oct 2010 10:20:19 -0400 (EDT) X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=-1.9 tagged_above=-10 required=5 tests=BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001 X-Spam-Level: X-Archive-Number: 201010/149 X-Sequence-Number: 40568 The discussions I've seen indicated that, in use, tablespaces were at th= e database level, but, yes, the docs do say that a table can be assigned= to a defined tablespace. What I still can't find is syntax which estab= lishes buffers/caches/whatever and assigns them to tablespaces. Without= that, I'm not sure what benefit there is to tablespaces, other than a s= ort of RAID-lite. Robert ---- Original message ---- >Date: Tue,= 12 Oct 2010 08:34:23 -0400 >From: pgsql-performance-owner@postgresql.o= rg (on behalf of Robert Haas ) >Subject: Re: [PE= RFORM] How does PG know if data is in memory? = >To: gnuoytr@rcn.com >Cc: pgsql-performance@postgresql.org > >On Mon,= Oct 11, 2010 at 11:11 PM, wrote: >> An approach tha= t works can be found in DB2, and likely elsewhere. >> >> The key is th= at tablespaces/tables/indexes/buffers are all attached through the buffe= rpool (the DB2 term). =A0A tablespace/bufferpool match is defined. =A0Th= en tables and indexes are assigned to the tablespace (and implicitly, th= e bufferpool). =A0As a result, one can effectively pin data in memory. =A0= This is very useful, but not low hanging fruit to implement. >> >> The= introduction of rudimentary tablespaces is a first step. =A0I assumed t= hat the point was to get to a DB2-like structure at some point. =A0Yes? > >We already have tablespaces, and our data already is accessed throug= h >the buffer pool. > >-- = >Robert Haas >EnterpriseDB: http://www.enterprisedb.com >The Enterpris= e PostgreSQL Company > >-- = >Sent via pgsql-performance mailing list (pgsql-performance@postgresql.o= rg) >To make changes to your subscription: >http://www.postgresql.org/= mailpref/pgsql-performance