<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="8" style="border: 1px #c9c399 solid;">
<tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://reviewboard.asterisk.org/r/2888/">https://reviewboard.asterisk.org/r/2888/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<p style="margin-top: 0;">On September 26th, 2013, 10:46 p.m. UTC, <b>Paul Belanger</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">"Not subscribing when not available buys a measurable amount of performance when running under load" is there any specific numbers you can add? Any measurement of performance impact documented will help when deploying asterisk in production.
</pre>
</blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Any numbers quoted would be highly dependent on the system it was run on, and hence kind of meaningless.
(1) To do the testing we've done over the past few days, we've done a SIPp scenario that issues calls at 25 cps to a dialplan that Answers, Waits(1), and hangs up. Unless someone executed that same scenario, they'd get wildly different results. Even with that same scenario, there's no guarantee you'll get similar results.
(2) Even with the same dialplan, SIPp scenario, and modules, you'll still get different real numbers. 10% load on one box could be 25% on another.
The real point is: both publishing and subscriptions aren't free. This patch deals with subscriptions; Josh has been working on publishing; David and Mark have been working on routing (and a whole mess of other areas). The act of creating, publishing, routing, and processing messages has a cost, and this patch drives down part of that cost when it doesn't need to be paid. That isn't terribly surprising, and - to an extent - has always been the case even in prior versions of Asterisk. For example, running a performance test with CDRs enabled (but not in batch mode) will yield much worse performance versus a system that has them disabled or has batch mode enabled. Part of that is simply the threading issues without batch mode; part is the processing that occurs when writing CDRs.</pre>
<br />
<p>- Matt</p>
<br />
<p>On September 26th, 2013, 8:18 p.m. UTC, Matt Jordan wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" style="background-image: url('https://reviewboard.asterisk.org/static/rb/images/review_request_box_top_bg.png'); background-position: left top; background-repeat: repeat-x; border: 1px black solid;">
<tr>
<td>
<div>Review request for Asterisk Developers, David Lee, Joshua Colp, and Mark Michelson.</div>
<div>By Matt Jordan.</div>
<p style="color: grey;"><i>Updated Sept. 26, 2013, 8:18 p.m.</i></p>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
Asterisk
</div>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: 1px solid #b8b5a0">
<tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Subscribing to Stasis isn't free.
As such, this patch makes AMI, CDR, and CEL - the "big 3" - only subscribe when enabled. Toggling their availability via a .conf file will unsubscribe/subscribe as appropriate.
Not subscribing when not available buys a measurable amount of performance when running under load.</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>/branches/12/main/cel.c <span style="color: grey">(399908)</span></li>
<li>/branches/12/main/cdr.c <span style="color: grey">(399908)</span></li>
<li>/branches/12/main/manager.c <span style="color: grey">(399908)</span></li>
</ul>
<p><a href="https://reviewboard.asterisk.org/r/2888/diff/" style="margin-left: 3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>