1
0
Fork 0
mirror of git://git.psyc.eu/libpsyc synced 2024-08-15 03:19:02 +00:00
libpsyc/bench/benchmark.html

1105 lines
32 KiB
HTML
Raw Normal View History

<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
<title>libpsyc Performance Benchmarks</title>
<!-- 2015-08-14 Fri 10:43 -->
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<meta name="generator" content="Org-mode" />
<style type="text/css">
<!--/*--><![CDATA[/*><!--*/
.title { text-align: center; }
.todo { font-family: monospace; color: red; }
.done { color: green; }
.tag { background-color: #eee; font-family: monospace;
padding: 2px; font-size: 80%; font-weight: normal; }
.timestamp { color: #bebebe; }
.timestamp-kwd { color: #5f9ea0; }
.right { margin-left: auto; margin-right: 0px; text-align: right; }
.left { margin-left: 0px; margin-right: auto; text-align: left; }
.center { margin-left: auto; margin-right: auto; text-align: center; }
.underline { text-decoration: underline; }
#postamble p, #preamble p { font-size: 90%; margin: .2em; }
p.verse { margin-left: 3%; }
pre {
border: 1px solid #ccc;
box-shadow: 3px 3px 3px #eee;
padding: 8pt;
font-family: monospace;
overflow: auto;
margin: 1.2em;
}
pre.src {
position: relative;
overflow: visible;
padding-top: 1.2em;
}
pre.src:before {
display: none;
position: absolute;
background-color: white;
top: -10px;
right: 10px;
padding: 3px;
border: 1px solid black;
}
pre.src:hover:before { display: inline;}
pre.src-sh:before { content: 'sh'; }
pre.src-bash:before { content: 'sh'; }
pre.src-emacs-lisp:before { content: 'Emacs Lisp'; }
pre.src-R:before { content: 'R'; }
pre.src-perl:before { content: 'Perl'; }
pre.src-java:before { content: 'Java'; }
pre.src-sql:before { content: 'SQL'; }
table { border-collapse:collapse; }
caption.t-above { caption-side: top; }
caption.t-bottom { caption-side: bottom; }
td, th { vertical-align:top; }
th.right { text-align: center; }
th.left { text-align: center; }
th.center { text-align: center; }
td.right { text-align: right; }
td.left { text-align: left; }
td.center { text-align: center; }
dt { font-weight: bold; }
.footpara:nth-child(2) { display: inline; }
.footpara { display: block; }
.footdef { margin-bottom: 1em; }
.figure { padding: 1em; }
.figure p { text-align: center; }
.inlinetask {
padding: 10px;
border: 2px solid gray;
margin: 10px;
background: #ffffcc;
}
#org-div-home-and-up
{ text-align: right; font-size: 70%; white-space: nowrap; }
textarea { overflow-x: auto; }
.linenr { font-size: smaller }
.code-highlighted { background-color: #ffff00; }
.org-info-js_info-navigation { border-style: none; }
#org-info-js_console-label
{ font-size: 10px; font-weight: bold; white-space: nowrap; }
.org-info-js_search-highlight
{ background-color: #ffff00; color: #000000; font-weight: bold; }
/*]]>*/-->
</style>
<script type="text/javascript">
/*
@licstart The following is the entire license notice for the
JavaScript code in this tag.
Copyright (C) 2012-2013 Free Software Foundation, Inc.
The JavaScript code in this tag is free software: you can
redistribute it and/or modify it under the terms of the GNU
General Public License (GNU GPL) as published by the Free Software
Foundation, either version 3 of the License, or (at your option)
any later version. The code is distributed WITHOUT ANY WARRANTY;
without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU GPL for more details.
As additional permission under GNU GPL version 3 section 7, you
may distribute non-source (e.g., minimized or compacted) forms of
that code without the copy of the GNU GPL normally required by
section 4, provided you include this license notice and a URL
through which recipients can access the Corresponding Source.
@licend The above is the entire license notice
for the JavaScript code in this tag.
*/
<!--/*--><![CDATA[/*><!--*/
function CodeHighlightOn(elem, id)
{
var target = document.getElementById(id);
if(null != target) {
elem.cacheClassElem = elem.className;
elem.cacheClassTarget = target.className;
target.className = "code-highlighted";
elem.className = "code-highlighted";
}
}
function CodeHighlightOff(elem, id)
{
var target = document.getElementById(id);
if(elem.cacheClassElem)
elem.className = elem.cacheClassElem;
if(elem.cacheClassTarget)
target.className = elem.cacheClassTarget;
}
/*]]>*///-->
</script>
<script type="text/javascript" src="http://orgmode.org/mathjax/MathJax.js"></script>
<script type="text/javascript">
<!--/*--><![CDATA[/*><!--*/
MathJax.Hub.Config({
// Only one of the two following lines, depending on user settings
// First allows browser-native MathML display, second forces HTML/CSS
// config: ["MMLorHTML.js"], jax: ["input/TeX"],
jax: ["input/TeX", "output/HTML-CSS"],
extensions: ["tex2jax.js","TeX/AMSmath.js","TeX/AMSsymbols.js",
"TeX/noUndefined.js"],
tex2jax: {
inlineMath: [ ["\\(","\\)"] ],
displayMath: [ ['$$','$$'], ["\\[","\\]"], ["\\begin{displaymath}","\\end{displaymath}"] ],
skipTags: ["script","noscript","style","textarea","pre","code"],
ignoreClass: "tex2jax_ignore",
processEscapes: false,
processEnvironments: true,
preview: "TeX"
},
showProcessingMessages: true,
displayAlign: "center",
displayIndent: "2em",
"HTML-CSS": {
scale: 100,
availableFonts: ["STIX","TeX"],
preferredFont: "TeX",
webFont: "TeX",
imageFont: "TeX",
showMathMenu: true,
},
MMLorHTML: {
prefer: {
MSIE: "MML",
Firefox: "MML",
Opera: "HTML",
other: "HTML"
}
}
});
/*]]>*///-->
</script>
</head>
<body>
<div id="content">
<h1 class="title">libpsyc Performance Benchmarks</h1>
<div id="table-of-contents">
<h2>Table of Contents</h2>
<div id="text-table-of-contents">
<ul>
<li><a href="#sec-1">1. PSYC, JSON, XML Syntax Benchmarks</a>
<ul>
<li><a href="#sec-1-1">1.1. User Profile</a></li>
<li><a href="#sec-1-2">1.2. A message with JSON-unfriendly characters</a></li>
<li><a href="#sec-1-3">1.3. A message with XML-unfriendly characters</a></li>
<li><a href="#sec-1-4">1.4. A message with PSYC-unfriendly strings</a></li>
<li><a href="#sec-1-5">1.5. Packets containing binary data</a></li>
</ul>
</li>
<li><a href="#sec-2">2. PSYC vs XMPP Protocol Benchmarks</a>
<ul>
<li><a href="#sec-2-1">2.1. A presence packet</a></li>
<li><a href="#sec-2-2">2.2. An average chat message</a></li>
<li><a href="#sec-2-3">2.3. A new status updated activity</a></li>
</ul>
</li>
<li><a href="#sec-3">3. Results</a></li>
<li><a href="#sec-4">4. Explanations</a></li>
<li><a href="#sec-5">5. Criticism</a></li>
<li><a href="#sec-6">6. Ease of Implementation</a></li>
<li><a href="#sec-7">7. Conclusions</a></li>
<li><a href="#sec-8">8. Futures</a></li>
<li><a href="#sec-9">9. Related Work</a></li>
<li><a href="#sec-10">10. Further Reading</a></li>
<li><a href="#sec-11">11. Appendix</a>
<ul>
<li><a href="#sec-11-1">11.1. Tools used</a></li>
</ul>
</li>
</ul>
</div>
</div>
<style type="text/css"> body { padding: 44px; } </style>
<p>
In this document we present the results of performance benchmarks
of libpsyc compared to json-c, libjson-glib, rapidxml and libxml2.
</p>
<div id="outline-container-sec-1" class="outline-2">
<h2 id="sec-1"><span class="section-number-2">1</span> PSYC, JSON, XML Syntax Benchmarks</h2>
<div class="outline-text-2" id="text-1">
<p>
First we look at the mere performance of the PSYC syntax
compared to equivalent XML and JSON encodings. We'll
look at actual XMPP messaging later.
</p>
</div>
<div id="outline-container-sec-1-1" class="outline-3">
<h3 id="sec-1-1"><span class="section-number-3">1.1</span> User Profile</h3>
<div class="outline-text-3" id="text-1-1">
<p>
In this test we'll compare the efficiency of the three
syntaxes at serializing a typical user data base
storage information. Let's start with XML:
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;UserProfile&gt;
&lt;Name&gt;Silvio Berlusconi&lt;/Name&gt;
&lt;JobTitle&gt;Premier&lt;/JobTitle&gt;
&lt;Country&gt;I&lt;/Country&gt;
&lt;Address&gt;
&lt;Street&gt;Via del Colosseo, 1&lt;/Street&gt;
&lt;PostalCode&gt;00100&lt;/PostalCode&gt;
&lt;City&gt;Roma&lt;/City&gt;
&lt;/Address&gt;
&lt;Page&gt;http://example.org&lt;/Page&gt;
&lt;/UserProfile&gt;
</pre>
</div>
<p>
In JSON this could look like this:
</p>
<div class="org-src-container">
<pre class="src src-js">["UserProfile",{"Name":"Silvio Berlusconi","JobTitle":"Premier","Country":"I","Address":
{"Street":"Via del Colosseo, 1","PostalCode":"00100","City":"Roma"},"Page":"http://example.org"}]
</pre>
</div>
<p>
Here's a way to model this in PSYC (verbose mode):
</p>
<div class="org-src-container">
<pre class="src src-psyc">:_name Silvio Berlusconi
:_title_job Premier
:_country I
:_address_street Via del Colosseo, 1
:_address_code_postal 00100
:_address_city Roma
:_page http://example.org
_profile_user
|
</pre>
</div>
</div>
</div>
<div id="outline-container-sec-1-2" class="outline-3">
<h3 id="sec-1-2"><span class="section-number-3">1.2</span> A message with JSON-unfriendly characters</h3>
<div class="outline-text-3" id="text-1-2">
<p>
This message contains some characters which are
impractical to encode in JSON. We should probably
put a lot more inside to actually see an impact
on performance. <b>TODO</b>
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;message from='romeo@example.net/orchard' to='juliet@example.com/balcony'&gt;
&lt;body&gt;"Neither, fair saint, if either thee dislike.", he said.
And
the
rest
is
history.&lt;/body&gt;
&lt;/message&gt;
</pre>
</div>
<div class="org-src-container">
<pre class="src src-js">["message",{"from":"romeo@example.net/orchard","to":"juliet@example.com/balcony"},
"\"Neither, fair saint, if either thee dislike.\", he said.\nAnd\nthe\nrest\nis\nhistory."]
</pre>
</div>
<div class="org-src-container">
<pre class="src src-psyc">:_source psyc://example.com/~romeo
:_target psyc://example.net/~juliet
_message
"Neither, fair saint, if either thee dislike.", he said.
And
the
rest
is
history.
|
</pre>
</div>
</div>
</div>
<div id="outline-container-sec-1-3" class="outline-3">
<h3 id="sec-1-3"><span class="section-number-3">1.3</span> A message with XML-unfriendly characters</h3>
<div class="outline-text-3" id="text-1-3">
<p>
Same test with characters which aren't practical
in the XML syntax, yet we should put more of
them inside. <b>TODO</b>
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;message from='juliet@example.com/balcony' to='romeo@example.net'&gt;
&lt;body&gt;Pro&amp;#x010D;e&amp;#x017D; jsi ty, Romeo?&lt;/body&gt;
&lt;/message&gt;
</pre>
</div>
<div class="org-src-container">
<pre class="src src-js">["message",{"from":"juliet@example.com/balcony","to":"romeo@example.net"},
"Pro\u010de\u017d jsi ty, Romeo?"]
</pre>
</div>
<div class="org-src-container">
<pre class="src src-psyc">:_source psyc://example.com/~juliet
:_target psyc://example.net/~romeo
_message
Pro&amp;#x010D;e&amp;#x017D; jsi ty, Romeo?
|
</pre>
</div>
</div>
</div>
<div id="outline-container-sec-1-4" class="outline-3">
<h3 id="sec-1-4"><span class="section-number-3">1.4</span> A message with PSYC-unfriendly strings</h3>
<div class="outline-text-3" id="text-1-4">
<p>
PSYC prefixes data with length as soon as it
exceeds certain sizes or contains certain strings.
In the case of short messages this is less
efficient than scanning the values without lengths.
Also, lengths are harder to edit by hand.
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;message from='juliet@example.com/balcony' to='romeo@example.net'&gt;
&lt;subject&gt;I implore you with a pointless
newline in a header variable&lt;/subject&gt;
&lt;body&gt;Wherefore art thou, Romeo?
|
And for practicing purposes we added a PSYC packet delimiter.&lt;/body&gt;
&lt;/message&gt;
</pre>
</div>
<div class="org-src-container">
<pre class="src src-js">["message",{"from":"juliet@example.com/balcony","to":"romeo@example.net",
"subject":"I implore you with a pointless\nnewline in a header variable"},
"Wherefore art thou, Romeo?\n|\nAnd for practicing purposes we added a PSYC packet delimiter."]
</pre>
</div>
<div class="org-src-container">
<pre class="src src-psyc">:_source psyc://example.com/~juliet
:_target psyc://example.net/~romeo
173
:_subject 59 I implore you with a pointless
newline in a header variable
_message
Wherefore art thou, Romeo?
|
And for practicing purposes we added a PSYC packet delimiter.
|
</pre>
</div>
</div>
</div>
<div id="outline-container-sec-1-5" class="outline-3">
<h3 id="sec-1-5"><span class="section-number-3">1.5</span> Packets containing binary data</h3>
<div class="outline-text-3" id="text-1-5">
<p>
We'll use a generator of random binary data to
see how well the formats behave with different
sizes of data. We'll consider 7000 as a possible
size of an icon, 70000 for an avatar, 700000
for a photograph, 7000000 for a piece of music,
70000000 for a large project and
700000000 for the contents of a CD.
</p>
</div>
</div>
</div>
<div id="outline-container-sec-2" class="outline-2">
<h2 id="sec-2"><span class="section-number-2">2</span> PSYC vs XMPP Protocol Benchmarks</h2>
<div class="outline-text-2" id="text-2">
<p>
These tests use typical messages from the XMPP ("stanzas" in
Jabber lingo) and compare them with equivalent JSON encodings
and PSYC formats.
</p>
</div>
<div id="outline-container-sec-2-1" class="outline-3">
<h3 id="sec-2-1"><span class="section-number-3">2.1</span> A presence packet</h3>
<div class="outline-text-3" id="text-2-1">
<p>
Since presence packets are by far the dominant messaging content
in the XMPP network, we'll start with one of them.
Here's an example from paragraph 4.4.2 of RFC 6121.
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;presence from='juliet@example.com/balcony'
to='benvolio@example.net'&gt;
&lt;show&gt;away&lt;/show&gt;
&lt;/presence&gt;
</pre>
</div>
<p>
And here's the same information in a JSON rendition:
</p>
<div class="org-src-container">
<pre class="src src-js">["presence",{"from":"juliet@example.com/balcony","to":"benvolio@example.net"},{"show":"away"}]
</pre>
</div>
<p>
Here's the equivalent PSYC packet in verbose mode
(since it is a multicast, the single recipients do not
need to be mentioned):
</p>
<div class="org-src-container">
<pre class="src src-psyc">:_context psyc://example.com/~juliet
=_degree_availability 4
_notice_presence
|
</pre>
</div>
<p>
And this is the same message in PSYC's compact form, but since compact mode
hasn't been implemented nor deployed yet, you should only consider this
for future projects:
</p>
<div class="org-src-container">
<pre class="src src-psyc">:c psyc://example.com/~juliet
=da 4
np
|
</pre>
</div>
</div>
</div>
<div id="outline-container-sec-2-2" class="outline-3">
<h3 id="sec-2-2"><span class="section-number-3">2.2</span> An average chat message</h3>
<div class="outline-text-3" id="text-2-2">
<div class="org-src-container">
<pre class="src src-xml">&lt;message from='juliet@example.com/balcony' to='romeo@example.net' type='chat'&gt;
&lt;body&gt;Art thou not Romeo, and a Montague?&lt;/body&gt;
&lt;/message&gt;
</pre>
</div>
<div class="org-src-container">
<pre class="src src-js">["message",{"from":"juliet@example.com/balcony","to":"romeo@example.net"},
"Art thou not Romeo, and a Montague?"]
</pre>
</div>
<div class="org-src-container">
<pre class="src src-psyc">:_source psyc://example.com/~juliet
:_target xmpp:romeo@example.net
_message
Art thou not Romeo, and a Montague?
|
</pre>
</div>
<p>
Little difference: PSYC by default doesn't mention a "resource" in XMPP terms,
instead it allows for more addressing schemes than just PSYC.
</p>
</div>
</div>
<div id="outline-container-sec-2-3" class="outline-3">
<h3 id="sec-2-3"><span class="section-number-3">2.3</span> A new status updated activity</h3>
<div class="outline-text-3" id="text-2-3">
<p>
Example taken from <a href="http://onesocialweb.org/spec/1.0/osw-activities.html">http://onesocialweb.org/spec/1.0/osw-activities.html</a>
You could call this XML namespace hell.. :-)
</p>
<div class="org-src-container">
<pre class="src src-xml">&lt;iq type='set'
from='hamlet@denmark.lit/snsclient'
to='hamlet@denmark.lit'
id='osw1'&gt;
&lt;pubsub xmlns='http://jabber.org/protocol/pubsub'&gt;
&lt;publish node='urn:xmpp:microblog:0'&gt;
&lt;item&gt;
&lt;entry xmlns="http://www.w3.org/2005/Atom"
xmlns:activity="http://activitystrea.ms/spec/1.0/"
xmlns:osw="http://onesocialweb.org/spec/1.0/"&gt;
&lt;title&gt;to be or not to be ?&lt;/title&gt;
&lt;activity:verb&gt;http://activitystrea.ms/schema/1.0/post&lt;/activity:verb&gt;
&lt;activity:object&gt;
&lt;activity:object-type&gt;http://onesocialweb.org/spec/1.0/object/status&lt;/activity:object-type&gt;
&lt;content type="text/plain"&gt;to be or not to be ?&lt;/content&gt;
&lt;/activity:object&gt;
&lt;osw:acl-rule&gt;
&lt;osw:acl-action permission="http://onesocialweb.org/spec/1.0/acl/permission/grant"&gt;
http://onesocialweb.org/spec/1.0/acl/action/view
&lt;/osw:acl-action&gt;
&lt;osw:acl-subject type="http://onesocialweb.org/spec/1.0/acl/subject/everyone"/&gt;
&lt;/osw:acl-rule&gt;
&lt;/entry&gt;
&lt;/item&gt;
&lt;/publish&gt;
&lt;/pubsub&gt;
&lt;/iq&gt;
</pre>
</div>
<p>
<a href="http://activitystrea.ms/head/json-activity.html">http://activitystrea.ms/head/json-activity.html</a> proposes a JSON encoding
of this. We'll have to add a routing header to it.
</p>
<div class="org-src-container">
<pre class="src src-js">["activity",{"from":"hamlet@denmark.lit/snsclient"},{"verb":"post",
"title":"to be or not to be ?","object":{"type":"status",
"content":"to be or not to be ?","contentType":"text/plain"}}]
</pre>
</div>
<p>
<a href="http://about.psyc.eu/Activity">http://about.psyc.eu/Activity</a> suggests a PSYC mapping for activity
streams. Should a "status post" be considered equivalent to a presence
description announcement or just a message in the "microblogging" channel?
We'll use the latter here:
</p>
<div class="org-src-container">
<pre class="src src-psyc">:_context psyc://denmark.lit/~hamlet#_follow
:_subject to be or not to be ?
:_type_content text/plain
_message
to be or not to be ?
|
</pre>
</div>
<p>
It's nice about XML namespaces how they can by definition never collide,
but this degree of engineering perfection causes us a lot of overhead.
The PSYC approach is to just extend the name of the method - as long as
people use differing method names, protocol extensions can exist next
to each other happily. Method name unicity cannot mathematically be ensured,
but it's enough to append your company name to make it unlikely for anyone
else on earth to have the same name. How this kind of safety is delivered
when using the JSON syntax of ActivityStreams is unclear. Apparently it was
no longer an important design criterion.
</p>
</div>
</div>
</div>
<div id="outline-container-sec-3" class="outline-2">
<h2 id="sec-3"><span class="section-number-2">3</span> Results</h2>
<div class="outline-text-2" id="text-3">
<p>
Parsing time of 1 000 000 packets, in milliseconds.
A simple strlen() scan of the respective message is provided for comparison.
These tests were performed on a 2.53 GHz Intel(R) Core(TM)2 Duo P9500 CPU.
</p>
<table border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<colgroup>
<col class="left" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
<col class="right" />
</colgroup>
<thead>
<tr>
<th scope="col" class="left">&#xa0;</th>
<th scope="col" class="right">strlen</th>
<th scope="col" class="right">libpsyc</th>
<th scope="col" class="right">json-c</th>
<th scope="col" class="right">json-glib</th>
<th scope="col" class="right">libxml sax</th>
<th scope="col" class="right">libxml</th>
<th scope="col" class="right">rapidxml</th>
</tr>
</thead>
<tbody>
<tr>
<td class="left">user profile</td>
<td class="right">55</td>
<td class="right">608</td>
<td class="right">4715</td>
<td class="right">16503</td>
<td class="right">7350</td>
<td class="right">12377</td>
<td class="right">2477</td>
</tr>
<tr>
<td class="left">psyc-unfriendly</td>
<td class="right">70</td>
<td class="right">286</td>
<td class="right">2892</td>
<td class="right">12567</td>
<td class="right">5538</td>
<td class="right">8659</td>
<td class="right">1896</td>
</tr>
<tr>
<td class="left">json-unfriendly</td>
<td class="right">49</td>
<td class="right">430</td>
<td class="right">2328</td>
<td class="right">10006</td>
<td class="right">5141</td>
<td class="right">7875</td>
<td class="right">1751</td>
</tr>
<tr>
<td class="left">xml-unfriendly</td>
<td class="right">37</td>
<td class="right">296</td>
<td class="right">2156</td>
<td class="right">9591</td>
<td class="right">5571</td>
<td class="right">8769</td>
<td class="right">1765</td>
</tr>
</tbody>
</table>
<p>
Pure syntax comparisons above, protocol performance comparisons below:
</p>
<table border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<colgroup>
<col class="left" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
<col class="center" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
<col class="right" />
</colgroup>
<thead>
<tr>
<th scope="col" class="left">&#xa0;</th>
<th scope="col" class="right">strlen</th>
<th scope="col" class="right">libpsyc</th>
<th scope="col" class="center">libpsyc compact</th>
<th scope="col" class="right">json-c</th>
<th scope="col" class="right">json-glib</th>
<th scope="col" class="right">libxml sax</th>
<th scope="col" class="right">libxml</th>
<th scope="col" class="right">rapidxml</th>
</tr>
</thead>
<tbody>
<tr>
<td class="left">presence</td>
<td class="right">30</td>
<td class="right">236</td>
<td class="center">122</td>
<td class="right">2463</td>
<td class="right">10016</td>
<td class="right">4997</td>
<td class="right">7557</td>
<td class="right">1719</td>
</tr>
<tr>
<td class="left">chat msg</td>
<td class="right">40</td>
<td class="right">295</td>
<td class="center">258</td>
<td class="right">2147</td>
<td class="right">9526</td>
<td class="right">5911</td>
<td class="right">8999</td>
<td class="right">1850</td>
</tr>
<tr>
<td class="left">activity</td>
<td class="right">42</td>
<td class="right">353</td>
<td class="center">279</td>
<td class="right">4666</td>
<td class="right">16327</td>
<td class="right">13357</td>
<td class="right">28858</td>
<td class="right">4356</td>
</tr>
</tbody>
</table>
<p>
Parsing large amounts of binary data. For JSON &amp; XML base64 encoding was used.
Note that the results below include only the parsing time, base64 decoding was
not performed.
</p>
<table border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<colgroup>
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
</colgroup>
<colgroup>
<col class="right" />
<col class="right" />
<col class="right" />
</colgroup>
<thead>
<tr>
<th scope="col" class="right">&#xa0;</th>
<th scope="col" class="right">strlen</th>
<th scope="col" class="right">libpsyc</th>
<th scope="col" class="right">json-c</th>
<th scope="col" class="right">json-glib</th>
<th scope="col" class="right">libxml sax</th>
<th scope="col" class="right">libxml</th>
<th scope="col" class="right">rapidxml</th>
</tr>
</thead>
<tbody>
<tr>
<td class="right">7K</td>
<td class="right">978</td>
<td class="right">77</td>
<td class="right">18609</td>
<td class="right">98000</td>
<td class="right">11445</td>
<td class="right">19299</td>
<td class="right">8701</td>
</tr>
<tr>
<td class="right">70K</td>
<td class="right">9613</td>
<td class="right">77</td>
<td class="right">187540</td>
<td class="right">1003900</td>
<td class="right">96209</td>
<td class="right">167738</td>
<td class="right">74296</td>
</tr>
<tr>
<td class="right">700K</td>
<td class="right">95888</td>
<td class="right">77</td>
<td class="right">1883500</td>
<td class="right">10616000</td>
<td class="right">842025</td>
<td class="right">1909428</td>
<td class="right">729419</td>
</tr>
<tr>
<td class="right">7M</td>
<td class="right">1347300</td>
<td class="right">78</td>
<td class="right">26359000</td>
<td class="right">120810000</td>
<td class="right">12466610</td>
<td class="right">16751363</td>
<td class="right">7581169</td>
</tr>
<tr>
<td class="right">70M</td>
<td class="right">14414000</td>
<td class="right">80</td>
<td class="right">357010000</td>
<td class="right">1241000000</td>
<td class="right">169622110</td>
<td class="right">296017820</td>
<td class="right">75308906</td>
</tr>
</tbody>
</table>
<p>
In each case we compared performance of parsing and re-rendering
these messages, but consider also that the applicative processing
of an XML DOM tree is more complicated than just accessing
certain elements in a JSON data structure or PSYC variable mapping.
</p>
</div>
</div>
<div id="outline-container-sec-4" class="outline-2">
<h2 id="sec-4"><span class="section-number-2">4</span> Explanations</h2>
<div class="outline-text-2" id="text-4">
<p>
As you can tell the PSYC data format outpaces its rivals in all circumstances.
Extremely so when delivering binary data as PSYC simply returns the starting
point and the length of the given buffer while the other parsers have to scan
for the end of the transmission, but also with many simpler operations, when
PSYC quickly figures out where the data starts and ends and passes such
information back to the application while the other formats are forced to
generate a copy of the data in order to process possibly embedded special
character sequences. PSYC essentially operates like a binary data protocol
even though it is actually text-based.
</p>
</div>
</div>
<div id="outline-container-sec-5" class="outline-2">
<h2 id="sec-5"><span class="section-number-2">5</span> Criticism</h2>
<div class="outline-text-2" id="text-5">
<p>
Are we comparing apples and oranges? Yes and no, depends on what you
need. XML is a syntax best suited for complex structured data in
well-defined formats - especially good for text mark-up. JSON is a syntax
intended to hold arbitrarily structured data suitable for immediate
inclusion in Javascript source codes. The PSYC syntax is an evolved
derivate of RFC 822, the syntax used by HTTP and E-Mail. It is currently
limited in the kind and depth of data structures that can be represented
with it, but it is highly efficient in exchange.
</p>
<p>
In fact we are currently looking into suitable syntax extensions to represent
generic structures and semantic signatures, but for now PSYC only
provides for simple typed values and lists of typed values.
</p>
</div>
</div>
<div id="outline-container-sec-6" class="outline-2">
<h2 id="sec-6"><span class="section-number-2">6</span> Ease of Implementation</h2>
<div class="outline-text-2" id="text-6">
<p>
Another aspect is the availability of these formats for spontaneous
use. You could generate and parse JSON yourself but you have to be
careful about escaping. XML can be rendered manually if you know your
data will not break the syntax, but you shouldn't dare to parse it without
a bullet proof parser. PSYC is easy to render and parse yourself for
simple tasks, as long as the body does not contain "\n|\n" and your
variables do not contain newlines.
</p>
</div>
</div>
<div id="outline-container-sec-7" class="outline-2">
<h2 id="sec-7"><span class="section-number-2">7</span> Conclusions</h2>
<div class="outline-text-2" id="text-7">
<p>
After all it is up to you to find out which format fulfils your
requirements the best. We use PSYC for the majority of messaging where
JSON and XMPP aren't efficient and opaque enough, but we employ XML and
JSON as payloads within PSYC for data that doesn't fit the PSYC model.
For some reason all three formats are being used for messaging, although
only PSYC was actually designed for that purpose.
</p>
<p>
The Internet has developed two major breeds of protocol formats.
The binary ones are extremely efficient but in most cases you have
to recompile all instances each time you change something
while the plain-text ones are reaching out for achieving perfection
in data representation while leaving the path of efficiency. Some
protocols such as HTTP and SIP are in-between these two schools,
offering both a text-based extensible syntax (it's actually easier to
add a header to HTTP than to come up with a namespace for XMPP&#x2026;)
and the ability to deliver binary data. But these protocols do not
come with native data structure support. PSYC is a protocol that
combines the compactness and efficiency of binary protocols with the
extensibility of text-based protocols and still provides for enough
data structuring to rarely require the use of other data formats.
</p>
</div>
</div>
<div id="outline-container-sec-8" class="outline-2">
<h2 id="sec-8"><span class="section-number-2">8</span> Futures</h2>
<div class="outline-text-2" id="text-8">
<p>
After a month of development libpsyc is already performing pretty
well, but we presume various optimizations, like rewriting parts
in assembler, are possible.
</p>
</div>
</div>
<div id="outline-container-sec-9" class="outline-2">
<h2 id="sec-9"><span class="section-number-2">9</span> Related Work</h2>
<div class="outline-text-2" id="text-9">
<p>
If this didn't help, you can also look into:
</p>
<ul class="org-ul">
<li>Adobe AMF
</li>
<li>ASN.1
</li>
<li>BSON
</li>
<li>Cisco Etch
</li>
<li>Efficient XML
</li>
<li>Facebook Thrift
</li>
<li>Google Protocol Buffers
</li>
</ul>
<p>
The drawback of these binary formats is, unlike PSYC, JSON and XML
you can't edit them manually and you can't produce valid messages
by replacing variables in a simple text template. You depend on
specialized parsers and renderers to be provided.
</p>
<p>
There's also
</p>
<ul class="org-ul">
<li>Bittorrent's bencode
</li>
</ul>
<p>
This format is formally text-based, but not easy to read as it doesn't
have any visual separators and isn't easy to edit as everything is
prefixed by lengths even for very short items.
</p>
</div>
</div>
<div id="outline-container-sec-10" class="outline-2">
<h2 id="sec-10"><span class="section-number-2">10</span> Further Reading</h2>
<div class="outline-text-2" id="text-10">
<p>
<a href="http://about.psyc.eu/Spec:Syntax">http://about.psyc.eu/Spec:Syntax</a> provides you with the ABNF grammar
of the PSYC 1.0 syntax. You may also be interested in PSYC's decentralized
state mechanism provided by the +/-/= operators.
</p>
<p>
See <a href="http://about.psyc.eu/XML">http://about.psyc.eu/XML</a> and <a href="http://about.psyc.eu/JSON">http://about.psyc.eu/JSON</a> for more
biased information on the respective formats.
</p>
</div>
</div>
<div id="outline-container-sec-11" class="outline-2">
<h2 id="sec-11"><span class="section-number-2">11</span> Appendix</h2>
<div class="outline-text-2" id="text-11">
</div><div id="outline-container-sec-11-1" class="outline-3">
<h3 id="sec-11-1"><span class="section-number-3">11.1</span> Tools used</h3>
<div class="outline-text-3" id="text-11-1">
<p>
This document and its benchmarks are distributed with libpsyc.
See <a href="http://about.psyc.eu/libpsyc">http://about.psyc.eu/libpsyc</a> on how to obtain it.
</p>
<p>
The benchmarks can be run with the following command
(xmlbench is needed for the xml tests):
</p>
<pre class="example">
make bench
</pre>
</div>
</div>
</div>
</div>
<div id="postamble" class="status">
<p class="date">Created: 2015-08-14 Fri 10:43</p>
<p class="creator"><a href="http://www.gnu.org/software/emacs/">Emacs</a> 24.4.1 (<a href="http://orgmode.org">Org</a> mode 8.2.6)</p>
<p class="validation"><a href="http://validator.w3.org/check?uri=referer">Validate</a></p>
</div>
</body>
</html>