Return-Path: X-Original-To: apmail-httpd-users-archive@www.apache.org Delivered-To: apmail-httpd-users-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id B613417C0E for ; Mon, 14 Sep 2015 05:22:23 +0000 (UTC) Received: (qmail 16297 invoked by uid 500); 14 Sep 2015 05:22:20 -0000 Delivered-To: apmail-httpd-users-archive@httpd.apache.org Received: (qmail 16254 invoked by uid 500); 14 Sep 2015 05:22:20 -0000 Mailing-List: contact users-help@httpd.apache.org; run by ezmlm Precedence: bulk Reply-To: users@httpd.apache.org list-help: list-unsubscribe: List-Post: List-Id: Delivered-To: mailing list users@httpd.apache.org Received: (qmail 16244 invoked by uid 99); 14 Sep 2015 05:22:19 -0000 Received: from Unknown (HELO spamd3-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 14 Sep 2015 05:22:19 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd3-us-west.apache.org (ASF Mail Server at spamd3-us-west.apache.org) with ESMTP id 7CB4A180424 for ; Mon, 14 Sep 2015 05:22:19 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd3-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: 4.001 X-Spam-Level: **** X-Spam-Status: No, score=4.001 tagged_above=-999 required=6.31 tests=[HTML_MESSAGE=3, KAM_LAZY_DOMAIN_SECURITY=1, URIBL_BLOCKED=0.001] autolearn=disabled Received: from mx1-us-east.apache.org ([10.40.0.8]) by localhost (spamd3-us-west.apache.org [10.40.0.10]) (amavisd-new, port 10024) with ESMTP id uBAhl8d39YYk for ; Mon, 14 Sep 2015 05:22:09 +0000 (UTC) Received: from mailgw.allianz.de (mailgw.allianz.de [194.127.3.35]) by mx1-us-east.apache.org (ASF Mail Server at mx1-us-east.apache.org) with ESMTPS id C815742B78 for ; Mon, 14 Sep 2015 05:22:08 +0000 (UTC) Received: from SW007429.wwg00m.rootdom.net (sw007429.wwg00m.rootdom.net [10.103.111.10] (may be forged)) by mailgw.allianz.de with ESMTP id t8E5Lmfw022312 for ; Mon, 14 Sep 2015 05:22:00 GMT Received: from WMUCV561.wwg00m.rootdom.net ([fe80::a9dc:6a8b:596f:b4de]) by SW007429.wwg00m.rootdom.net ([fe80::b8e8:bd7f:448d:694c%13]) with mapi; Mon, 14 Sep 2015 07:21:47 +0200 From: "Bremser, Kurt (AMOS Austria GmbH)" To: "users@httpd.apache.org" Date: Mon, 14 Sep 2015 07:17:54 +0200 Thread-Topic: [users@httpd] Memory mapping in Apache question [wd-vc] Thread-Index: AQHQ7q1AuoB+JcjrE0WjzHgE2U7Ndw== Message-ID: <02D980EB3A9003459E275CA771F3F4820141992B2679@WMUCV561.wwg00m.rootdom.net> References: In-Reply-To: Accept-Language: de-DE Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: de-DE Content-Type: multipart/alternative; boundary="_000_02D980EB3A9003459E275CA771F3F4820141992B2679WMUCV561wwg_" MIME-Version: 1.0 Subject: Re: [users@httpd] Memory mapping in Apache question [wd-vc] --_000_02D980EB3A9003459E275CA771F3F4820141992B2679WMUCV561wwg_ Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 7bit On any serious operating system, accessing the same content will be memory-mapped after the first request, since from then on the file(s) in question will be cached. Reducing the complexity inside apache may even give a benefit. Kurt Bremser AMOS Austria Newton was wrong. There is no gravity. The Earth sucks. ________________________________ Von: Rose, John B [jbrose@utk.edu] Gesendet: Freitag, 11. September 2015 22:28 An: users@httpd.apache.org Betreff: **SPAM?** [users@httpd] Memory mapping in Apache question [wd-vc] Apache 2.2 Prefork Red Hat 6 We did a small test with a single image to compare results of it as memory mapped content and not memory mapped. I expected the memory mapped image to be quicker to load. But in our testing that did not seem to be the case. Should accessing a memory mapped image be quicker than non-memory mapped? I'm doing the below ab test to compare the performance, each 3 times: ab -n 100000 -c 20 http://abc.com/ If we disable MMAP, It takes 14.555 seconds, 12.775 seconds, 10.698 seconds. Average: 12.675s. If we enable MMAP, It takes 18.468 seconds, 13.882 seconds, 16.532 seconds Average: 16.294s. Thanks AMOS Austria GmbH 1130 Wien, Hietzinger Kai 101-105 FN 365014k, Handelsgericht Wien UID: ATU 66614737 http://www.allianz.at ******************************************************** Dieses E-Mail und allfaellig daran angeschlossene Anhaenge enthalten Informationen, die vertraulich und ausschliesslich fuer den (die) bezeichneten Adressaten bestimmt sind. Wenn Sie nicht der genannte Adressat sind, darf dieses E-Mail samt allfaelliger Anhaenge von Ihnen weder anderen Personen zugaenglich gemacht noch in anderer Weise verwertet werden. Wenn Sie nicht der beabsichtigte Empfaenger sind, bitten wir Sie, dieses E-Mail und saemtliche angeschlossene Anhaenge zu loeschen. Please note: This email and any files transmitted with it is intended only for the named recipients and may contain confidential and/or privileged information. If you are not the intended recipient, please do not read, copy, use or disclose the contents of this communication to others and notify the sender immediately. Then please delete the email and any copies of it. Thank you. ******************************************************** --_000_02D980EB3A9003459E275CA771F3F4820141992B2679WMUCV561wwg_ Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 7bit
On any serious operating system, accessing the same content will be memory-mapped after the first request, since from then on the file(s) in question will be cached.
Reducing the complexity inside apache may even give a benefit.
 
Kurt Bremser
AMOS Austria
 
Newton was wrong. There is no gravity. The Earth sucks.
Von: Rose, John B [jbrose@utk.edu]
Gesendet: Freitag, 11. September 2015 22:28
An: users@httpd.apache.org
Betreff: **SPAM?** [users@httpd] Memory mapping in Apache question [wd-vc]

Apache 2.2
Prefork
Red Hat 6

We did a small test with a single image to compare results of it as memory mapped content and not memory mapped. I expected the memory mapped image to be quicker to load. But in our testing that did not seem to be the case. Should accessing a memory mapped image be quicker than non-memory mapped?

I'm doing the below ab test to compare the performance, each 3 times:

ab -n 100000 -c 20 http://abc.com/


If we disable MMAP,

It takes 14.555 seconds, 12.775 seconds, 10.698 seconds.

Average: 12.675s.


If we enable MMAP,
It takes 18.468 seconds, 13.882 seconds, 16.532 seconds

Average: 16.294s.



Thanks

AMOS Austria GmbH
1130 Wien, Hietzinger Kai 101-105
FN 365014k, Handelsgericht Wien
UID: ATU 66614737

http://www.allianz.at

********************************************************
Dieses E-Mail und allfaellig daran angeschlossene Anhaenge
enthalten Informationen, die vertraulich und
ausschliesslich fuer den (die) bezeichneten Adressaten
bestimmt sind.
Wenn Sie nicht der genannte Adressat sind, darf dieses
E-Mail samt allfaelliger Anhaenge von Ihnen weder anderen
Personen zugaenglich gemacht noch in anderer Weise
verwertet werden.
Wenn Sie nicht der beabsichtigte Empfaenger sind, bitten
wir Sie, dieses E-Mail und saemtliche angeschlossene
Anhaenge zu loeschen.

Please note: This email and any files transmitted with it is
intended only for the named recipients and may contain
confidential and/or privileged information. If you are not the
intended recipient, please do not read, copy, use or disclose
the contents of this communication to others and notify the
sender immediately. Then please delete the email and any
copies of it. Thank you.
******************************************************** --_000_02D980EB3A9003459E275CA771F3F4820141992B2679WMUCV561wwg_--