Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@httpd.apache.org
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: mod_cache performance
Date: Tue, 3 Aug 2004 10:11:05 -0700
Message-ID: 
 <16A24B6050584143978000646801E9350201D54C@cacexc03.americas.cpqcorp.net>
Thread-Topic: mod_cache performance
Thread-Index: AcR5aLeH9gCkVUYOQTK36CpS3bLnTAAEHCHA
From: "Mathihalli, Madhusudan" <madhum@hp.com>
To: <dev@httpd.apache.org>


: -----Original Message-----
: From: Bill Stoddard [mailto:bill@wstoddard.com]=20
[SNIP]
: >=20
: > Here's some comparative numbers to chew on.
: >=20
: > One client and one server on 100Mbps network (cheapy=20
: 100Base-T switch);
: > 50 simulated users hitting 7 URLs 100 times with flood=20
: (35,000 requests).
: >=20
: > mod_disk_cache: Requests: 35000 Time: 40.91 Req/Sec: 856.78
: > mod_mem_cache:  Requests: 35000 Time: 54.90 Req/Sec: 637.81
: > no cache:       Requests: 35000 Time: 54.86 Req/Sec: 638.81
: > squid:          Requests: 35000 Time: 105.35 Req/Sec: 332.25
: >=20
: > mod_disk_cache completely filled out the network at ~50% CPU usage.
: >    [Can't push through more than ~8MB/sec (~64Mb/sec) without GigE.]
: > mod_mem_cache filled up the CPU but not the network
: >    [Poor scaling characteristics.  It goes to 100% CPU with=20
: just 5 users!]
:=20
: mod_mem_cache is broken then. It used to kick the pants off=20
: of 'no cache' and mod_disk_cache.

.. Well, doesn't it depend upon the size of the data set. With 'ab', I
guess that's possible that mod_mem_cache can beat mod_disk_cache - but
with a dataset like SPECweb99, I'd really doubt if it can really do it.

BTW, I wonder how mem_cache can significantly out-perform no-cache
scenario - 'cause a good file system should buffer cache the
most-accessed files, and there should be minimal perf. difference.

-Madhu