From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx197.postini.com [74.125.245.197]) by kanga.kvack.org (Postfix) with SMTP id 943076B0033 for ; Thu, 25 Jul 2013 11:07:14 -0400 (EDT) From: =?iso-8859-1?q?C=E9dric_Villemain?= Reply-To: cedric@2ndquadrant.com Subject: Re: mincore() & fincore() Date: Thu, 25 Jul 2013 17:07:10 +0200 References: <201307251658.33548.cedric@2ndquadrant.com> In-Reply-To: <201307251658.33548.cedric@2ndquadrant.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="Boundary-01=_e8T8R0HEwjRLbF3" Content-Transfer-Encoding: 7bit Message-Id: <201307251707.11159.cedric@2ndquadrant.com> Sender: owner-linux-mm@kvack.org List-ID: To: linux-mm@kvack.org Cc: Johannes Weiner --Boundary-01=_e8T8R0HEwjRLbF3 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable [sorry, previous mail was sent earlier than expected] > First, the proposed changes in this email are to be used at least for=20 > PostgreSQL extensions, maybe for core. >=20 > Purpose is to offer better monitoring/tracking of the hot/cold areas (and= =20 > read/write paterns) in the tables and indexes, in PostgreSQL those are by= default=20 > written in segments of 1GB. >=20 > There are some possible usecase already: >=20 > * planning of hardware upgrade > * easier configuration setup (both PostgreSQL and linux) > * provide more informations to the planner/executor of PostgreSQL >=20 > My ideas so far are to=20 >=20 > * improve mincore() in linux and add it information like in freeBSD (at= =20 > least adding 'mincore_modified' to track clean vs dirty pages). > * adding fincore() to make the information easier to grab from PostgreSQ= L (no=20 > mmap) > * maybe some access to those stats in /proc/ >=20 > It makes years that libprefetch, mincore() and fincore() are discussed on= linux=20 > mailling lists. And they got a good feedback... So I hope it is ok to kee= p on=20 > those and provide updated patches. Johannes, I add you in CC because you're the last one who proposed somethin= g.=20 Can I update your patch with previous suggestions from reviewers ? I'm also asking for feedback in this area, others ideas are very welcome. =2D-=20 C=E9dric Villemain +33 (0)6 20 30 22 52 http://2ndQuadrant.fr/ PostgreSQL: Support 24x7 - D=E9veloppement, Expertise et Formation --Boundary-01=_e8T8R0HEwjRLbF3 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

[sorry, pre= vious mail was sent earlier than expected]

&nb= sp;

> First,= the proposed changes in this email are to be used at least for

> Postgr= eSQL extensions, maybe for core.

> Purpos= e is to offer better monitoring/tracking of the hot/cold areas (and

> read/w= rite paterns) in the tables and indexes, in PostgreSQL those are by default=

> writte= n in segments of 1GB.

> There = are some possible usecase already:

> * pla= nning of hardware upgrade

> * eas= ier configuration setup (both PostgreSQL and linux)

> * pro= vide more informations to the planner/executor of PostgreSQL

> My ide= as so far are to

> * imp= rove mincore() in linux and add it information like in freeBSD (at

> lea= st adding 'mincore_modified' to track clean vs dirty pages).

> * add= ing fincore() to make the information easier to grab from PostgreSQL (no

> mma= p)

> * may= be some access to those stats in /proc/

> It mak= es years that libprefetch, mincore() and fincore() are discussed on linux <= /p>

> mailli= ng lists. And they got a good feedback... So I hope it is ok to keep on

> those = and provide updated patches.

&nb= sp;

Johannes, I= add you in CC because you're the last one who proposed something.

Can I updat= e your patch with previous suggestions from reviewers ?

&nb= sp;

I'm also as= king for feedback in this area, others ideas are very welcome.

&nb= sp;

C=E9dric Vi= llemain +33 (0)6 20 30 22 52

http://2ndQ= uadrant.fr/

PostgreSQL:= Support 24x7 - D=E9veloppement, Expertise et Formation

&nb= sp;

--Boundary-01=_e8T8R0HEwjRLbF3-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org