<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="#0563C1" vlink="#954F72">
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt">I want to confirm if this behavior is to be expected…<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">I have built a custom CentOS RPM (based on previous threads to use 1.9.3) since we just want to use unbound for forwarding and not do any validation. What I found is that the performance is significantly worse
than the GA CentOS 7 RPM (1.6.6) with validation enabled.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">We are testing resolution of 2000 domains. On the RPM we built the average is roughly 50 domains per second. The config only uses unbound as a forwarder:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">server:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> username: "unbound"<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> pidfile: "/var/run/unbound.pid"<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> verbosity: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> interface: 127.0.0.1<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> do-ip6: no<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> access-control: 127.0.0.0/8 allow<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> domain-insecure: "."<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> module-config: "iterator"<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> prefetch: yes<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> num-threads: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> outgoing-range: 8192<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> num-queries-per-thread: 4096<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> so-reuseport: yes<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> msg-cache-slabs: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> rrset-cache-slabs: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> infra-cache-slabs: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"> key-cache-slabs: 2<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">On the GA RPM (using the same config file) but with the following 2 options commented out:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">domain-insecure: "."<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">module-config: "iterator"<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">The performance is much better. We are able to do roughly 400 domains per second. I tried commenting them out on the 1.9.3 deployment and there was no noticeable performance impact.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">My main question is the behavior expected? I was under the impression that non-validating would be faster than a validating server.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">As stated before the main reason we are trying the custom build is to use unbound only for forwarding and not do root server look ups.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:10.0pt;font-family:"Arial",sans-serif;color:black">Thanks,<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:10.0pt;font-family:"Arial",sans-serif;color:black">Daniel<o:p></o:p></span></p>
</div>
</body>
</html>