xforms

[apache] / docs / manual / misc / perf-tuning.xml

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE manualpage SYSTEM "../style/manualpage.dtd">
<?xml-stylesheet type="text/xsl" href="../style/manual.en.xsl"?>
<!-- $LastChangedRevision$ -->

<!--
 Licensed to the Apache Software Foundation (ASF) under one or more
 contributor license agreements.  See the NOTICE file distributed with
 this work for additional information regarding copyright ownership.
 The ASF licenses this file to You under the Apache License, Version 2.0
 (the "License"); you may not use this file except in compliance with
 the License.  You may obtain a copy of the License at

     http://www.apache.org/licenses/LICENSE-2.0

 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
-->

<manualpage metafile="perf-tuning.xml.meta">
  <parentdocument href="./">Miscellaneous Documentation</parentdocument>

  <title>Apache Performance Tuning</title>

  <summary>

    <p>Apache 2.x is a general-purpose webserver, designed to
    provide a balance of flexibility, portability, and performance.
    Although it has not been designed specifically to set benchmark
    records, Apache 2.x is capable of high performance in many
    real-world situations.</p>

    <p>Compared to Apache 1.3, release 2.x contains many additional
    optimizations to increase throughput and scalability. Most of
    these improvements are enabled by default. However, there are
    compile-time and run-time configuration choices that can
    significantly affect performance. This document describes the
    options that a server administrator can configure to tune the
    performance of an Apache 2.x installation. Some of these
    configuration options enable the httpd to better take advantage
    of the capabilities of the hardware and OS, while others allow
    the administrator to trade functionality for speed.</p>

  </summary>

  <section id="hardware">

    <title>Hardware and Operating System Issues</title>

    <p>The single biggest hardware issue affecting webserver
    performance is RAM. A webserver should never ever have to swap,
    as swapping increases the latency of each request beyond a point
    that users consider "fast enough". This causes users to hit
    stop and reload, further increasing the load. You can, and
    should, control the <directive module="mpm_common"
    >MaxRequestWorkers</directive> setting so that your server
    does not spawn so many children it starts swapping. This procedure
    for doing this is simple: determine the size of your average Apache
    process, by looking at your process list via a tool such as
    <code>top</code>, and divide this into your total available memory,
    leaving some room for other processes.</p>

    <p>Beyond that the rest is mundane: get a fast enough CPU, a
    fast enough network card, and fast enough disks, where "fast
    enough" is something that needs to be determined by
    experimentation.</p>

    <p>Operating system choice is largely a matter of local
    concerns. But some guidelines that have proven generally
    useful are:</p>

    <ul>
      <li>
        <p>Run the latest stable release and patchlevel of the
        operating system that you choose. Many OS suppliers have
        introduced significant performance improvements to their
        TCP stacks and thread libraries in recent years.</p>
      </li>

      <li>
        <p>If your OS supports a <code>sendfile(2)</code> system
        call, make sure you install the release and/or patches
        needed to enable it. (With Linux, for example, this means
        using Linux 2.4 or later. For early releases of Solaris 8,
        you may need to apply a patch.) On systems where it is
        available, <code>sendfile</code> enables Apache 2 to deliver
        static content faster and with lower CPU utilization.</p>
      </li>
    </ul>

  </section>

  <section id="runtime">

    <title>Run-Time Configuration Issues</title>

    <related>
      <modulelist>
        <module>mod_dir</module>
        <module>mpm_common</module>
        <module>mod_status</module>
      </modulelist>
      <directivelist>
        <directive module="core">AllowOverride</directive>
        <directive module="mod_dir">DirectoryIndex</directive>
        <directive module="core">HostnameLookups</directive>
        <directive module="core">EnableMMAP</directive>
        <directive module="core">EnableSendfile</directive>
        <directive module="core">KeepAliveTimeout</directive>
        <directive module="prefork">MaxSpareServers</directive>
        <directive module="prefork">MinSpareServers</directive>
        <directive module="core">Options</directive>
        <directive module="mpm_common">StartServers</directive>
      </directivelist>
    </related>

    <section id="dns">

      <title>HostnameLookups and other DNS considerations</title>

      <p>Prior to Apache 1.3, <directive module="core"
      >HostnameLookups</directive> defaulted to <code>On</code>.
      This adds latency to every request because it requires a
      DNS lookup to complete before the request is finished. In
      Apache 1.3 this setting defaults to <code>Off</code>. If you need
      to have addresses in your log files resolved to hostnames, use the
      <program>logresolve</program>
      program that comes with Apache, or one of the numerous log
      reporting packages which are available.</p>

      <p>It is recommended that you do this sort of postprocessing of
      your log files on some machine other than the production web
      server machine, in order that this activity not adversely affect
      server performance.</p>

      <p>If you use any <code><directive module="mod_access_compat"
      >Allow</directive> from domain</code> or <code><directive
      module="mod_access_compat">Deny</directive> from domain</code>
      directives (i.e., using a hostname, or a domain name, rather than
      an IP address) then you will pay for
      two DNS lookups (a reverse, followed by a forward lookup
      to make sure that the reverse is not being spoofed). For best
      performance, therefore, use IP addresses, rather than names, when
      using these directives, if possible.</p>

      <p>Note that it's possible to scope the directives, such as
      within a <code>&lt;Location /server-status&gt;</code> section.
      In this case the DNS lookups are only performed on requests
      matching the criteria. Here's an example which disables lookups
      except for <code>.html</code> and <code>.cgi</code> files:</p>

      <highlight language="config">
HostnameLookups off
&lt;Files ~ "\.(html|cgi)$"&gt;
  HostnameLookups on
&lt;/Files&gt;
      </highlight>

      <p>But even still, if you just need DNS names in some CGIs you
      could consider doing the <code>gethostbyname</code> call in the
      specific CGIs that need it.</p>

    </section>

    <section id="symlinks">

      <title>FollowSymLinks and SymLinksIfOwnerMatch</title>

      <p>Wherever in your URL-space you do not have an <code>Options
      FollowSymLinks</code>, or you do have an <code>Options
      SymLinksIfOwnerMatch</code> Apache will have to issue extra
      system calls to check up on symlinks. One extra call per
      filename component. For example, if you had:</p>

      <highlight language="config">
DocumentRoot /www/htdocs
&lt;Directory /&gt;
  Options SymLinksIfOwnerMatch
&lt;/Directory&gt;
      </highlight>

      <p>and a request is made for the URI <code>/index.html</code>.
      Then Apache will perform <code>lstat(2)</code> on
      <code>/www</code>, <code>/www/htdocs</code>, and
      <code>/www/htdocs/index.html</code>. The results of these
      <code>lstats</code> are never cached, so they will occur on
      every single request. If you really desire the symlinks
      security checking you can do something like this:</p>

      <highlight language="config">
DocumentRoot /www/htdocs
&lt;Directory /&gt;
  Options FollowSymLinks
&lt;/Directory&gt;

&lt;Directory /www/htdocs&gt;
  Options -FollowSymLinks +SymLinksIfOwnerMatch
&lt;/Directory&gt;
      </highlight>

      <p>This at least avoids the extra checks for the
      <directive module="core">DocumentRoot</directive> path.
      Note that you'll need to add similar sections if you
      have any <directive module="mod_alias">Alias</directive> or
      <directive module="mod_rewrite">RewriteRule</directive> paths
      outside of your document root. For highest performance,
      and no symlink protection, set <code>FollowSymLinks</code>
      everywhere, and never set <code>SymLinksIfOwnerMatch</code>.</p>

    </section>

    <section id="htaccess">

      <title>AllowOverride</title>

      <p>Wherever in your URL-space you allow overrides (typically
      <code>.htaccess</code> files) Apache will attempt to open
      <code>.htaccess</code> for each filename component. For
      example,</p>

      <highlight language="config">
DocumentRoot /www/htdocs
&lt;Directory /&gt;
  AllowOverride all
&lt;/Directory&gt;
      </highlight>

      <p>and a request is made for the URI <code>/index.html</code>.
      Then Apache will attempt to open <code>/.htaccess</code>,
      <code>/www/.htaccess</code>, and
      <code>/www/htdocs/.htaccess</code>. The solutions are similar
      to the previous case of <code>Options FollowSymLinks</code>.
      For highest performance use <code>AllowOverride None</code>
      everywhere in your filesystem.</p>

    </section>

    <section id="negotiation">

      <title>Negotiation</title>

      <p>If at all possible, avoid content-negotiation if you're
      really interested in every last ounce of performance. In
      practice the benefits of negotiation outweigh the performance
      penalties. There's one case where you can speed up the server.
      Instead of using a wildcard such as:</p>

      <highlight language="config">DirectoryIndex index</highlight>

      <p>Use a complete list of options:</p>

      <highlight language="config">DirectoryIndex index.cgi index.pl index.shtml index.html</highlight>

      <p>where you list the most common choice first.</p>

      <p>Also note that explicitly creating a <code>type-map</code>
      file provides better performance than using
      <code>MultiViews</code>, as the necessary information can be
      determined by reading this single file, rather than having to
      scan the directory for files.</p>

    <p>If your site needs content negotiation consider using
    <code>type-map</code> files, rather than the <code>Options
    MultiViews</code> directive to accomplish the negotiation. See the
    <a href="../content-negotiation.html">Content Negotiation</a>
    documentation for a full discussion of the methods of negotiation,
    and instructions for creating <code>type-map</code> files.</p>

    </section>

    <section>

      <title>Memory-mapping</title>

      <p>In situations where Apache 2.x needs to look at the contents
      of a file being delivered--for example, when doing server-side-include
      processing--it normally memory-maps the file if the OS supports
      some form of <code>mmap(2)</code>.</p>

      <p>On some platforms, this memory-mapping improves performance.
      However, there are cases where memory-mapping can hurt the performance
      or even the stability of the httpd:</p>

      <ul>
        <li>
          <p>On some operating systems, <code>mmap</code> does not scale
          as well as <code>read(2)</code> when the number of CPUs increases.
          On multiprocessor Solaris servers, for example, Apache 2.x sometimes
          delivers server-parsed files faster when <code>mmap</code> is disabled.</p>
        </li>

        <li>
          <p>If you memory-map a file located on an NFS-mounted filesystem
          and a process on another NFS client machine deletes or truncates
          the file, your process may get a bus error the next time it tries
          to access the mapped file content.</p>
        </li>
      </ul>

      <p>For installations where either of these factors applies, you
      should use <code>EnableMMAP off</code> to disable the memory-mapping
      of delivered files. (Note: This directive can be overridden on
      a per-directory basis.)</p>

    </section>

    <section>

      <title>Sendfile</title>

      <p>In situations where Apache 2.x can ignore the contents of the file
      to be delivered -- for example, when serving static file content --
      it normally uses the kernel sendfile support the file if the OS
      supports the <code>sendfile(2)</code> operation.</p>

      <p>On most platforms, using sendfile improves performance by eliminating
      separate read and send mechanics.  However, there are cases where using
      sendfile can harm the stability of the httpd:</p>

      <ul>
        <li>
          <p>Some platforms may have broken sendfile support that the build
          system did not detect, especially if the binaries were built on
          another box and moved to such a machine with broken sendfile support.</p>
        </li>
        <li>
          <p>With an NFS-mounted filesystem, the kernel may be unable
          to reliably serve the network file through its own cache.</p>
        </li>
      </ul>

      <p>For installations where either of these factors applies, you
      should use <code>EnableSendfile off</code> to disable sendfile
      delivery of file contents. (Note: This directive can be overridden
      on a per-directory basis.)</p>

    </section>

    <section id="process">

      <title>Process Creation</title>

      <p>Prior to Apache 1.3 the <directive module="prefork"
      >MinSpareServers</directive>, <directive module="prefork"
      >MaxSpareServers</directive>, and <directive module="mpm_common"
      >StartServers</directive> settings all had drastic effects on
      benchmark results. In particular, Apache required a "ramp-up"
      period in order to reach a number of children sufficient to serve
      the load being applied. After the initial spawning of
      <directive module="mpm_common">StartServers</directive> children,
      only one child per second would be created to satisfy the
      <directive module="prefork">MinSpareServers</directive>
      setting. So a server being accessed by 100 simultaneous
      clients, using the default <directive module="mpm_common"
      >StartServers</directive> of <code>5</code> would take on
      the order 95 seconds to spawn enough children to handle
      the load. This works fine in practice on real-life servers,
      because they aren't restarted frequently. But does really
      poorly on benchmarks which might only run for ten minutes.</p>

      <p>The one-per-second rule was implemented in an effort to
      avoid swamping the machine with the startup of new children. If
      the machine is busy spawning children it can't service
      requests. But it has such a drastic effect on the perceived
      performance of Apache that it had to be replaced. As of Apache
      1.3, the code will relax the one-per-second rule. It will spawn
      one, wait a second, then spawn two, wait a second, then spawn
      four, and it will continue exponentially until it is spawning
      32 children per second. It will stop whenever it satisfies the
      <directive module="prefork">MinSpareServers</directive>
      setting.</p>

      <p>This appears to be responsive enough that it's almost
      unnecessary to twiddle the <directive module="prefork"
      >MinSpareServers</directive>, <directive module="prefork"
      >MaxSpareServers</directive> and <directive module="mpm_common"
      >StartServers</directive> knobs. When more than 4 children are
      spawned per second, a message will be emitted to the
      <directive module="core">ErrorLog</directive>. If you
      see a lot of these errors then consider tuning these settings.
      Use the <module>mod_status</module> output as a guide.</p>

    <p>Related to process creation is process death induced by the
    <directive module="mpm_common">MaxConnectionsPerChild</directive>
    setting. By default this is <code>0</code>,
    which means that there is no limit to the number of connections
    handled per child. If your configuration currently has this set
    to some very low number, such as <code>30</code>, you may want to bump this
    up significantly. If you are running SunOS or an old version of
    Solaris, limit this to <code>10000</code> or so because of memory leaks.</p>

    <p>When keep-alives are in use, children will be kept busy
    doing nothing waiting for more requests on the already open
    connection. The default <directive module="core"
    >KeepAliveTimeout</directive> of <code>5</code>
    seconds attempts to minimize this effect. The tradeoff here is
    between network bandwidth and server resources. In no event
    should you raise this above about <code>60</code> seconds, as <a
    href="http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-95-4.html">
    most of the benefits are lost</a>.</p>

    </section>

  </section>

  <section id="compiletime">

    <title>Compile-Time Configuration Issues</title>

    <section>

      <title>Choosing an MPM</title>

      <p>Apache 2.x supports pluggable concurrency models, called
      <a href="../mpm.html">Multi-Processing Modules</a> (MPMs).
      When building Apache, you must choose an MPM to use.  There
      are platform-specific MPMs for some platforms:
      <module>mpm_netware</module>,
      <module>mpmt_os2</module>, and <module>mpm_winnt</module>.  For
      general Unix-type systems, there are several MPMs from which
      to choose.  The choice of MPM can affect the speed and scalability
      of the httpd:</p>

      <ul>

        <li>The <module>worker</module> MPM uses multiple child
        processes with many threads each.  Each thread handles
        one connection at a time.  Worker generally is a good
        choice for high-traffic servers because it has a smaller
        memory footprint than the prefork MPM.</li>
        
        <li>The <module>event</module> MPM is threaded like the 
        Worker MPM, but is designed to allow more requests to be 
        served simultaneously by passing off some processing work 
        to supporting threads, freeing up the main threads to work
        on new requests.</li>

        <li>The <module>prefork</module> MPM uses multiple child
        processes with one thread each.  Each process handles
        one connection at a time.  On many systems, prefork is
        comparable in speed to worker, but it uses more memory.
        Prefork's threadless design has advantages over worker
        in some situations: it can be used with non-thread-safe
        third-party modules, and it is easier to debug on platforms
        with poor thread debugging support.</li>

      </ul>

      <p>For more information on these and other MPMs, please
      see the MPM <a href="../mpm.html">documentation</a>.</p>

    </section>

    <section id="modules">

        <title>Modules</title>

        <p>Since memory usage is such an important consideration in
        performance, you should attempt to eliminate modules that you are
        not actually using. If you have built the modules as <a
        href="../dso.html">DSOs</a>, eliminating modules is a simple
        matter of commenting out the associated <directive
        module="mod_so">LoadModule</directive> directive for that module.
        This allows you to experiment with removing modules, and seeing
        if your site still functions in their absence.</p>

        <p>If, on the other hand, you have modules statically linked
        into your Apache binary, you will need to recompile Apache in
        order to remove unwanted modules.</p>

        <p>An associated question that arises here is, of course, what
        modules you need, and which ones you don't. The answer here
        will, of course, vary from one web site to another. However, the
        <em>minimal</em> list of modules which you can get by with tends
        to include <module>mod_mime</module>, <module>mod_dir</module>,
        and <module>mod_log_config</module>. <code>mod_log_config</code> is,
        of course, optional, as you can run a web site without log
        files. This is, however, not recommended.</p>

    </section>

    <section>

      <title>Atomic Operations</title>

      <p>Some modules, such as <module>mod_cache</module> and
      recent development builds of the worker MPM, use APR's
      atomic API.  This API provides atomic operations that can
      be used for lightweight thread synchronization.</p>

      <p>By default, APR implements these operations using the
      most efficient mechanism available on each target
      OS/CPU platform.  Many modern CPUs, for example, have
      an instruction that does an atomic compare-and-swap (CAS)
      operation in hardware.  On some platforms, however, APR
      defaults to a slower, mutex-based implementation of the
      atomic API in order to ensure compatibility with older
      CPU models that lack such instructions.  If you are
      building Apache for one of these platforms, and you plan
      to run only on newer CPUs, you can select a faster atomic
      implementation at build time by configuring Apache with
      the <code>--enable-nonportable-atomics</code> option:</p>

      <example>
        ./buildconf<br />
        ./configure --with-mpm=worker --enable-nonportable-atomics=yes
      </example>

      <p>The <code>--enable-nonportable-atomics</code> option is
      relevant for the following platforms:</p>

      <ul>

        <li>Solaris on SPARC<br />
            By default, APR uses mutex-based atomics on Solaris/SPARC.
            If you configure with <code>--enable-nonportable-atomics</code>,
            however, APR generates code that uses a SPARC v8plus opcode for
            fast hardware compare-and-swap.  If you configure Apache with
            this option, the atomic operations will be more efficient
            (allowing for lower CPU utilization and higher concurrency),
            but the resulting executable will run only on UltraSPARC
            chips.
        </li>

        <li>Linux on x86<br />
            By default, APR uses mutex-based atomics on Linux.  If you
            configure with <code>--enable-nonportable-atomics</code>,
            however, APR generates code that uses a 486 opcode for fast
            hardware compare-and-swap.  This will result in more efficient
            atomic operations, but the resulting executable will run only
            on 486 and later chips (and not on 386).
        </li>

      </ul>

    </section>

    <section>

      <title>mod_status and ExtendedStatus On</title>

      <p>If you include <module>mod_status</module> and you also set
      <code>ExtendedStatus On</code> when building and running
      Apache, then on every request Apache will perform two calls to
      <code>gettimeofday(2)</code> (or <code>times(2)</code>
      depending on your operating system), and (pre-1.3) several
      extra calls to <code>time(2)</code>. This is all done so that
      the status report contains timing indications. For highest
      performance, set <code>ExtendedStatus off</code> (which is the
      default).</p>

    </section>

    <section>

      <title>accept Serialization - multiple sockets</title>

    <note type="warning"><title>Warning:</title>
      <p>This section has not been fully updated
      to take into account changes made in the 2.x version of the
      Apache HTTP Server. Some of the information may still be
      relevant, but please use it with care.</p>
    </note>

      <p>This discusses a shortcoming in the Unix socket API. Suppose
      your web server uses multiple <directive module="mpm_common"
      >Listen</directive> statements to listen on either multiple
      ports or multiple addresses. In order to test each socket
      to see if a connection is ready Apache uses
      <code>select(2)</code>. <code>select(2)</code> indicates that a
      socket has <em>zero</em> or <em>at least one</em> connection
      waiting on it. Apache's model includes multiple children, and
      all the idle ones test for new connections at the same time. A
      naive implementation looks something like this (these examples
      do not match the code, they're contrived for pedagogical
      purposes):</p>

      <highlight language="c">
        for (;;) {
          for (;;) {
            fd_set accept_fds;

            FD_ZERO (&amp;accept_fds);
            for (i = first_socket; i &lt;= last_socket; ++i) {
              FD_SET (i, &amp;accept_fds);
            }
            rc = select (last_socket+1, &amp;accept_fds, NULL, NULL, NULL);
            if (rc &lt; 1) continue;
            new_connection = -1;
            for (i = first_socket; i &lt;= last_socket; ++i) {
              if (FD_ISSET (i, &amp;accept_fds)) {
                new_connection = accept (i, NULL, NULL);
                if (new_connection != -1) break;
              }
            }
            if (new_connection != -1) break;
          }
          process_the(new_connection);
        }
      </highlight>

      <p>But this naive implementation has a serious starvation problem.
      Recall that multiple children execute this loop at the same
      time, and so multiple children will block at
      <code>select</code> when they are in between requests. All
      those blocked children will awaken and return from
      <code>select</code> when a single request appears on any socket
      (the number of children which awaken varies depending on the
      operating system and timing issues). They will all then fall
      down into the loop and try to <code>accept</code> the
      connection. But only one will succeed (assuming there's still
      only one connection ready), the rest will be <em>blocked</em>
      in <code>accept</code>. This effectively locks those children
      into serving requests from that one socket and no other
      sockets, and they'll be stuck there until enough new requests
      appear on that socket to wake them all up. This starvation
      problem was first documented in <a
      href="http://bugs.apache.org/index/full/467">PR#467</a>. There
      are at least two solutions.</p>

      <p>One solution is to make the sockets non-blocking. In this
      case the <code>accept</code> won't block the children, and they
      will be allowed to continue immediately. But this wastes CPU
      time. Suppose you have ten idle children in
      <code>select</code>, and one connection arrives. Then nine of
      those children will wake up, try to <code>accept</code> the
      connection, fail, and loop back into <code>select</code>,
      accomplishing nothing. Meanwhile none of those children are
      servicing requests that occurred on other sockets until they
      get back up to the <code>select</code> again. Overall this
      solution does not seem very fruitful unless you have as many
      idle CPUs (in a multiprocessor box) as you have idle children,
      not a very likely situation.</p>

      <p>Another solution, the one used by Apache, is to serialize
      entry into the inner loop. The loop looks like this
      (differences highlighted):</p>

      <highlight language="c">
        for (;;) {
          <strong>accept_mutex_on ();</strong>
          for (;;) {
            fd_set accept_fds;
            
            FD_ZERO (&amp;accept_fds);
            for (i = first_socket; i &lt;= last_socket; ++i) {
              FD_SET (i, &amp;accept_fds);
            }
            rc = select (last_socket+1, &amp;accept_fds, NULL, NULL, NULL);
            if (rc &lt; 1) continue;
            new_connection = -1;
            for (i = first_socket; i &lt;= last_socket; ++i) {
              if (FD_ISSET (i, &amp;accept_fds)) {
                new_connection = accept (i, NULL, NULL);
                if (new_connection != -1) break;
              }
            }
            if (new_connection != -1) break;
          }
          <strong>accept_mutex_off ();</strong>
          process the new_connection;
        }
      </highlight>

      <p><a id="serialize" name="serialize">The functions</a>
      <code>accept_mutex_on</code> and <code>accept_mutex_off</code>
      implement a mutual exclusion semaphore. Only one child can have
      the mutex at any time. There are several choices for
      implementing these mutexes. The choice is defined in
      <code>src/conf.h</code> (pre-1.3) or
      <code>src/include/ap_config.h</code> (1.3 or later). Some
      architectures do not have any locking choice made, on these
      architectures it is unsafe to use multiple
      <directive module="mpm_common">Listen</directive>
      directives.</p>

      <p>The <directive module="core">Mutex</directive> directive can
      be used to change the mutex implementation of the
      <code>mpm-accept</code> mutex at run-time.  Special considerations
      for different mutex implementations are documented with that
      directive.</p>

      <p>Another solution that has been considered but never
      implemented is to partially serialize the loop -- that is, let
      in a certain number of processes. This would only be of
      interest on multiprocessor boxes where it's possible multiple
      children could run simultaneously, and the serialization
      actually doesn't take advantage of the full bandwidth. This is
      a possible area of future investigation, but priority remains
      low because highly parallel web servers are not the norm.</p>

      <p>Ideally you should run servers without multiple
      <directive module="mpm_common">Listen</directive>
      statements if you want the highest performance.
      But read on.</p>

    </section>

    <section>

      <title>accept Serialization - single socket</title>

      <p>The above is fine and dandy for multiple socket servers, but
      what about single socket servers? In theory they shouldn't
      experience any of these same problems because all children can
      just block in <code>accept(2)</code> until a connection
      arrives, and no starvation results. In practice this hides
      almost the same "spinning" behaviour discussed above in the
      non-blocking solution. The way that most TCP stacks are
      implemented, the kernel actually wakes up all processes blocked
      in <code>accept</code> when a single connection arrives. One of
      those processes gets the connection and returns to user-space,
      the rest spin in the kernel and go back to sleep when they
      discover there's no connection for them. This spinning is
      hidden from the user-land code, but it's there nonetheless.
      This can result in the same load-spiking wasteful behaviour
      that a non-blocking solution to the multiple sockets case
      can.</p>

      <p>For this reason we have found that many architectures behave
      more "nicely" if we serialize even the single socket case. So
      this is actually the default in almost all cases. Crude
      experiments under Linux (2.0.30 on a dual Pentium pro 166
      w/128Mb RAM) have shown that the serialization of the single
      socket case causes less than a 3% decrease in requests per
      second over unserialized single-socket. But unserialized
      single-socket showed an extra 100ms latency on each request.
      This latency is probably a wash on long haul lines, and only an
      issue on LANs. If you want to override the single socket
      serialization you can define
      <code>SINGLE_LISTEN_UNSERIALIZED_ACCEPT</code> and then
      single-socket servers will not serialize at all.</p>

    </section>

    <section>

      <title>Lingering Close</title>

      <p>As discussed in <a
      href="http://www.ics.uci.edu/pub/ietf/http/draft-ietf-http-connection-00.txt">
      draft-ietf-http-connection-00.txt</a> section 8, in order for
      an HTTP server to <strong>reliably</strong> implement the
      protocol it needs to shutdown each direction of the
      communication independently (recall that a TCP connection is
      bi-directional, each half is independent of the other).</p>

      <p>When this feature was added to Apache it caused a flurry of
      problems on various versions of Unix because of a
      shortsightedness. The TCP specification does not state that the
      <code>FIN_WAIT_2</code> state has a timeout, but it doesn't prohibit it.
      On systems without the timeout, Apache 1.2 induces many sockets
      stuck forever in the <code>FIN_WAIT_2</code> state. In many cases this
      can be avoided by simply upgrading to the latest TCP/IP patches
      supplied by the vendor. In cases where the vendor has never
      released patches (<em>i.e.</em>, SunOS4 -- although folks with
      a source license can patch it themselves) we have decided to
      disable this feature.</p>

      <p>There are two ways of accomplishing this. One is the socket
      option <code>SO_LINGER</code>. But as fate would have it, this
      has never been implemented properly in most TCP/IP stacks. Even
      on those stacks with a proper implementation (<em>i.e.</em>,
      Linux 2.0.31) this method proves to be more expensive (cputime)
      than the next solution.</p>

      <p>For the most part, Apache implements this in a function
      called <code>lingering_close</code> (in
      <code>http_main.c</code>). The function looks roughly like
      this:</p>

      <highlight language="c">
        void lingering_close (int s)
        {
          char junk_buffer[2048];
          
          /* shutdown the sending side */
          shutdown (s, 1);

          signal (SIGALRM, lingering_death);
          alarm (30);

          for (;;) {
            select (s for reading, 2 second timeout);
            if (error) break;
            if (s is ready for reading) {
              if (read (s, junk_buffer, sizeof (junk_buffer)) &lt;= 0) {
                break;
              }
              /* just toss away whatever is here */
            }
          }
          
          close (s);
        }
      </highlight>

      <p>This naturally adds some expense at the end of a connection,
      but it is required for a reliable implementation. As HTTP/1.1
      becomes more prevalent, and all connections are persistent,
      this expense will be amortized over more requests. If you want
      to play with fire and disable this feature you can define
      <code>NO_LINGCLOSE</code>, but this is not recommended at all.
      In particular, as HTTP/1.1 pipelined persistent connections
      come into use <code>lingering_close</code> is an absolute
      necessity (and <a
      href="http://www.w3.org/Protocols/HTTP/Performance/Pipeline.html">
      pipelined connections are faster</a>, so you want to support
      them).</p>

    </section>

    <section>

      <title>Scoreboard File</title>

      <p>Apache's parent and children communicate with each other
      through something called the scoreboard. Ideally this should be
      implemented in shared memory. For those operating systems that
      we either have access to, or have been given detailed ports
      for, it typically is implemented using shared memory. The rest
      default to using an on-disk file. The on-disk file is not only
      slow, but it is unreliable (and less featured). Peruse the
      <code>src/main/conf.h</code> file for your architecture and
      look for either <code>USE_MMAP_SCOREBOARD</code> or
      <code>USE_SHMGET_SCOREBOARD</code>. Defining one of those two
      (as well as their companions <code>HAVE_MMAP</code> and
      <code>HAVE_SHMGET</code> respectively) enables the supplied
      shared memory code. If your system has another type of shared
      memory, edit the file <code>src/main/http_main.c</code> and add
      the hooks necessary to use it in Apache. (Send us back a patch
      too please.)</p>

      <note>Historical note: The Linux port of Apache didn't start to
      use shared memory until version 1.2 of Apache. This oversight
      resulted in really poor and unreliable behaviour of earlier
      versions of Apache on Linux.</note>

    </section>

    <section>

      <title>DYNAMIC_MODULE_LIMIT</title>

      <p>If you have no intention of using dynamically loaded modules
      (you probably don't if you're reading this and tuning your
      server for every last ounce of performance) then you should add
      <code>-DDYNAMIC_MODULE_LIMIT=0</code> when building your
      server. This will save RAM that's allocated only for supporting
      dynamically loaded modules.</p>

    </section>

  </section>

  <section id="trace">

    <title>Appendix: Detailed Analysis of a Trace</title>

    <p>Here is a system call trace of Apache 2.0.38 with the worker MPM
    on Solaris 8. This trace was collected using:</p>

    <example>
      truss -l -p <var>httpd_child_pid</var>.
    </example>

    <p>The <code>-l</code> option tells truss to log the ID of the
    LWP (lightweight process--Solaris' form of kernel-level thread)
    that invokes each system call.</p>

    <p>Other systems may have different system call tracing utilities
    such as <code>strace</code>, <code>ktrace</code>, or <code>par</code>.
    They all produce similar output.</p>

    <p>In this trace, a client has requested a 10KB static file
    from the httpd. Traces of non-static requests or requests
    with content negotiation look wildly different (and quite ugly
    in some cases).</p>

    <example>
<pre>/67:    accept(3, 0x00200BEC, 0x00200C0C, 1) (sleeping...)
/67:    accept(3, 0x00200BEC, 0x00200C0C, 1)            = 9</pre>
    </example>

    <p>In this trace, the listener thread is running within LWP #67.</p>

    <note>Note the lack of <code>accept(2)</code> serialization. On this
    particular platform, the worker MPM uses an unserialized accept by
    default unless it is listening on multiple ports.</note>

    <example>
<pre>/65:    lwp_park(0x00000000, 0)                         = 0
/67:    lwp_unpark(65, 1)                               = 0</pre>
    </example>

    <p>Upon accepting the connection, the listener thread wakes up
    a worker thread to do the request processing. In this trace,
    the worker thread that handles the request is mapped to LWP #65.</p>

    <example>
<pre>/65:    getsockname(9, 0x00200BA4, 0x00200BC4, 1)       = 0</pre>
    </example>

    <p>In order to implement virtual hosts, Apache needs to know
    the local socket address used to accept the connection. It
    is possible to eliminate this call in many situations (such
    as when there are no virtual hosts, or when
    <directive module="mpm_common">Listen</directive> directives
    are used which do not have wildcard addresses). But
    no effort has yet been made to do these optimizations. </p>

    <example>
<pre>/65:    brk(0x002170E8)                                 = 0
/65:    brk(0x002190E8)                                 = 0</pre>
    </example>

    <p>The <code>brk(2)</code> calls allocate memory from the heap.
    It is rare to see these in a system call trace, because the httpd
    uses custom memory allocators (<code>apr_pool</code> and
    <code>apr_bucket_alloc</code>) for most request processing.
    In this trace, the httpd has just been started, so it must
    call <code>malloc(3)</code> to get the blocks of raw memory
    with which to create the custom memory allocators.</p>

    <example>
<pre>/65:    fcntl(9, F_GETFL, 0x00000000)                   = 2
/65:    fstat64(9, 0xFAF7B818)                          = 0
/65:    getsockopt(9, 65535, 8192, 0xFAF7B918, 0xFAF7B910, 2190656) = 0
/65:    fstat64(9, 0xFAF7B818)                          = 0
/65:    getsockopt(9, 65535, 8192, 0xFAF7B918, 0xFAF7B914, 2190656) = 0
/65:    setsockopt(9, 65535, 8192, 0xFAF7B918, 4, 2190656) = 0
/65:    fcntl(9, F_SETFL, 0x00000082)                   = 0</pre>
    </example>

    <p>Next, the worker thread puts the connection to the client (file
    descriptor 9) in non-blocking mode. The <code>setsockopt(2)</code>
    and <code>getsockopt(2)</code> calls are a side-effect of how
    Solaris' libc handles <code>fcntl(2)</code> on sockets.</p>

    <example>
<pre>/65:    read(9, " G E T   / 1 0 k . h t m".., 8000)     = 97</pre>
    </example>

    <p>The worker thread reads the request from the client.</p>

    <example>
<pre>/65:    stat("/var/httpd/apache/httpd-8999/htdocs/10k.html", 0xFAF7B978) = 0
/65:    open("/var/httpd/apache/httpd-8999/htdocs/10k.html", O_RDONLY) = 10</pre>
    </example>

    <p>This httpd has been configured with <code>Options FollowSymLinks</code>
    and <code>AllowOverride None</code>.  Thus it doesn't need to
    <code>lstat(2)</code> each directory in the path leading up to the
    requested file, nor check for <code>.htaccess</code> files.
    It simply calls <code>stat(2)</code> to verify that the file:
    1) exists, and 2) is a regular file, not a directory.</p>

    <example>
<pre>/65:    sendfilev(0, 9, 0x00200F90, 2, 0xFAF7B53C)      = 10269</pre>
    </example>

    <p>In this example, the httpd is able to send the HTTP response
    header and the requested file with a single <code>sendfilev(2)</code>
    system call. Sendfile semantics vary among operating systems. On some other
    systems, it is necessary to do a <code>write(2)</code> or
    <code>writev(2)</code> call to send the headers before calling
    <code>sendfile(2)</code>.</p>

    <example>
<pre>/65:    write(4, " 1 2 7 . 0 . 0 . 1   -  ".., 78)      = 78</pre>
    </example>

    <p>This <code>write(2)</code> call records the request in the
    access log. Note that one thing missing from this trace is a
    <code>time(2)</code> call. Unlike Apache 1.3, Apache 2.x uses
    <code>gettimeofday(3)</code> to look up the time. On some operating
    systems, like Linux or Solaris, <code>gettimeofday</code> has an
    optimized implementation that doesn't require as much overhead
    as a typical system call.</p>

    <example>
<pre>/65:    shutdown(9, 1, 1)                               = 0
/65:    poll(0xFAF7B980, 1, 2000)                       = 1
/65:    read(9, 0xFAF7BC20, 512)                        = 0
/65:    close(9)                                        = 0</pre>
    </example>

    <p>The worker thread does a lingering close of the connection.</p>

    <example>
<pre>/65:    close(10)                                       = 0
/65:    lwp_park(0x00000000, 0)         (sleeping...)</pre>
    </example>

    <p>Finally the worker thread closes the file that it has just delivered
    and blocks until the listener assigns it another connection.</p>

    <example>
<pre>/67:    accept(3, 0x001FEB74, 0x001FEB94, 1) (sleeping...)</pre>
    </example>

    <p>Meanwhile, the listener thread is able to accept another connection
    as soon as it has dispatched this connection to a worker thread (subject
    to some flow-control logic in the worker MPM that throttles the listener
    if all the available workers are busy).  Though it isn't apparent from
    this trace, the next <code>accept(2)</code> can (and usually does, under
    high load conditions) occur in parallel with the worker thread's handling
    of the just-accepted connection.</p>

  </section>

</manualpage>