Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@httpd.apache.org
Received-SPF: pass (herse.apache.org: local policy)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Subject: Re: mod_disk_cache summarization
Date: Tue, 24 Oct 2006 16:43:54 +0200
Message-ID: 
 <F6E88684A95CDB4DBBD5B9F4C1C4B09B020EB5B4@vis01-mx02.vis.internal.vodafone.com>
Thread-Topic: Re: mod_disk_cache summarization
Thread-Index: Acb3efuGvvon95fqTCSUIInZD8qbPgAAGB5w
From: =?iso-8859-1?Q?Pl=FCm=2C_R=FCdiger=2C_VF_EITO?=
 <ruediger.pluem@vodafone.com>
To: <dev@httpd.apache.org>


> -----Urspr=FCngliche Nachricht-----
> Von: Brian Akins =20
> Gesendet: Dienstag, 24. Oktober 2006 16:37
> An: dev@httpd.apache.org
> Betreff: Re: mod_disk_cache summarization
>=20
>=20
> Niklas Edmundsson wrote:
> >> The comparison of your and Brian's experience are two ends=20
> of extremes on
> >> high volume caches, one low hits large files, the second=20
> high hits small
> >> files. This should make for some useful tuning information.
> >=20
> > The extreme difference is what makes me think that we=20
> should acknowledge=20
> > that they exist and provide the relevant knobs where=20
> necessary. As it=20
> > looks right now, those knobs tend to be more OS/filesystem=20
> specific, but=20
> > that might change as this evolves.
>=20
>=20
> My thought on this is that we use providers, so in theory,=20
> you could use a=20
> different provider for the different types:
>=20
> CacheEnable /largecrap large_disk_with_stat_sleep_thing
> CacheEnable /normalstuff normal_disk

Agreed. If it turns out that the common code base between both
cases is only small and it is complex to do both things in one
provider just make two providers out of them. The remaining common
code could be factored out in a separate disk_cache_util c file
which is used by both providers.

Regards

R=FCdiger