<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=gb2312" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 9.00.8112.16434">
<STYLE>@font-face {
font-family: 宋体;
}
@font-face {
font-family: Verdana;
}
@font-face {
font-family: @宋体;
}
@page Section1 {size: 595.3pt 841.9pt; margin: 72.0pt 90.0pt 72.0pt 90.0pt; layout-grid: 15.6pt; }
P.MsoNormal {
TEXT-JUSTIFY: inter-ideograph; TEXT-ALIGN: justify; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman"; FONT-SIZE: 10.5pt
}
LI.MsoNormal {
TEXT-JUSTIFY: inter-ideograph; TEXT-ALIGN: justify; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman"; FONT-SIZE: 10.5pt
}
DIV.MsoNormal {
TEXT-JUSTIFY: inter-ideograph; TEXT-ALIGN: justify; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman"; FONT-SIZE: 10.5pt
}
A:link {
COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle17 {
FONT-STYLE: normal; FONT-FAMILY: Verdana; COLOR: windowtext; FONT-WEIGHT: normal; TEXT-DECORATION: none; mso-style-type: personal-compose
}
DIV.Section1 {
page: Section1
}
UNKNOWN {
FONT-SIZE: 10pt
}
BLOCKQUOTE {
MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px; MARGIN-LEFT: 2em
}
OL {
MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
UL {
MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</STYLE>
</HEAD>
<BODY style="MARGIN: 10px; FONT-FAMILY: verdana; FONT-SIZE: 10pt">
<DIV><FONT color=#000080 size=2 face=Verdana><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><FONT color=#000000>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>Dear
Wouter,<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office"
/><o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>Thank you
very much for your answer.<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>I set the
TTL to 0 in order to test the processing ability of DNS
softwares.<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>In this way,
all the queries should do : root query---TLD query---SLD query. No cache for all
queries and in all cases.<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><o:p><FONT
size=2><STRONG> </STRONG></FONT></o:p></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>When the TLD
server uses NSD, the performance is promoted than the all-BIND case. That means
the NSD can improve the DNS processing efficiency. However, because the SLD uses
BIND, the bottleneck still exists. So, when all the authority servers (both TLD
and SLD) use NSD. The performance is promoted
significantly.<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><o:p><FONT
size=2><STRONG> </STRONG></FONT></o:p></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>However,
when the SLD server uses NSD, the performance is not promoted than the all-BIND
case. So the results are different for the following two cases and that is just
my confusion.<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><o:p><FONT
size=2><STRONG> </STRONG></FONT></o:p></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>Case2:
Recursive (bind) ---Root (bind) ---TA (bind) ---SA
(nsd)<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>173.912569
qps<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>Case3:
Recursive (bind) --- Root (bind) ---TA (nsd) ---SA
(bind)<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><FONT size=2><STRONG>1857.768915
qps<o:p></o:p></STRONG></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><o:p><FONT
size=2><STRONG> </STRONG></FONT></o:p></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="FONT-FAMILY: 'Verdana','sans-serif'; mso-fareast-language: ZH-CN; mso-bidi-font-family: 宋体"
lang=EN-US><FONT size=2>You said: “The recursor must ask the TLD server to fetch
the NS records again. This is why performance depends on the TLD server.”
<o:p></o:p></FONT></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN><STRONG><FONT size=2>Yes,
but the recursor also must ask the SLD server to fetch the answer. So why NSD
cannot improve the performance of SLD
server?<o:p></o:p></FONT></STRONG></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><o:p><STRONG><FONT
size=2> </FONT></STRONG></o:p></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><STRONG><FONT
size=2>BR,<o:p></o:p></FONT></STRONG></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal><SPAN
style="mso-fareast-language: ZH-CN" lang=EN-US><STRONG><FONT size=2>Zhiwei
<o:p></o:p></FONT></STRONG></SPAN></P>
<P style="MARGIN: 0cm 0cm 0pt" class=MsoNormal></FONT></FONT></SPAN></FONT><FONT
color=#000080 face=Verdana></FONT></P></DIV>
<DIV><FONT color=#000080 face=Verdana></FONT> </DIV>
<DIV><FONT color=#c0c0c0 face=Verdana>2011-09-16 </FONT></DIV><FONT
color=#000080 face=Verdana>
<HR style="WIDTH: 100px" align=left color=#b5c4df SIZE=1>
</FONT>
<DIV><FONT color=#c0c0c0 face=Verdana><SPAN>yzw_iplab</SPAN> </FONT></DIV>
<HR color=#b5c4df SIZE=1>
<DIV><FONT face=Verdana><STRONG>发件人：</STRONG> W.C.A. Wijngaards </FONT></DIV>
<DIV><FONT face=Verdana><STRONG>发送时间：</STRONG> 2011-09-16 16:27:29
</FONT></DIV>
<DIV><FONT face=Verdana><STRONG>收件人：</STRONG> nsd-users </FONT></DIV>
<DIV><FONT face=Verdana><STRONG>抄送：</STRONG> </FONT></DIV>
<DIV><FONT face=Verdana><STRONG>主题：</STRONG> Re: [nsd-users] NSD test
</FONT></DIV>
<DIV><FONT face=Verdana></FONT> </DIV>
<DIV><FONT face=Verdana>
<DIV>-----BEGIN PGP SIGNED MESSAGE-----</DIV>
<DIV>Hash: SHA1</DIV>
<DIV></DIV>
<DIV>Hi Zhiwei Yan,</DIV>
<DIV></DIV>
<DIV>I do not fully understand your test. But I think that the 0 TTL affects</DIV>
<DIV>the NS records for my.test. and thus the recursor is not allowed to</DIV>
<DIV>cache the NS records. The recursor must ask the TLD server to fetch the</DIV>
<DIV>NS records again. This is why performance depends on the TLD server.</DIV>
<DIV></DIV>
<DIV>Best regards,</DIV>
<DIV> Wouter</DIV>
<DIV></DIV>
<DIV>On 09/16/2011 10:07 AM, yzw_iplab wrote:</DIV>
<DIV>> Hi, all,</DIV>
<DIV>> </DIV>
<DIV>> Recently, I tested the performance of BIND,NSD and UNBOUND. However, I</DIV>
<DIV>> am confused with the testing results. I list them here for your help.</DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> I fake the root, ‘test’ top level domain and ‘my.test’ second level domain.</DIV>
<DIV>> </DIV>
<DIV>> I use the ‘queryperf’ in the client and the number of queries is 50000,</DIV>
<DIV>> all of them belong to ‘my.test’ domain and have positive responses. In</DIV>
<DIV>> the zone file of ‘test’ and ‘my.test’, TTL is set to 0, and then every</DIV>
<DIV>> query should be performed from root to the ‘my.test’ server without the</DIV>
<DIV>> assistance of cache.</DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> Client executes the DNS query using ‘queryperf’ under the following</DIV>
<DIV>> four scenarios (where SA and TA denote the Second level Authority server</DIV>
<DIV>> and Top level Authority server respectively), their QPS results are：</DIV>
<DIV>> </DIV>
<DIV>> Case1: Recursive (bind)---Root(bind)---TA*(bind)*---*SA(bind)*</DIV>
<DIV>> </DIV>
<DIV>> 153.521955 qps</DIV>
<DIV>> </DIV>
<DIV>> Case2: Recursive(bind)---Root (bind)---TA*(bind)*---*SA (nsd)*</DIV>
<DIV>> </DIV>
<DIV>> 173.912569 qps</DIV>
<DIV>> </DIV>
<DIV>> Case3: Recursive(bind)---Root (bind)---TA*(nsd)*---*SA (bind)*</DIV>
<DIV>> </DIV>
<DIV>> 1857.768915 qps</DIV>
<DIV>> </DIV>
<DIV>> Case4: Recursive(bind)---Root (bind)---A*(nsd)*---*SA (nsd)*</DIV>
<DIV>> </DIV>
<DIV>> 1923.845425 qps</DIV>
<DIV>> </DIV>
<DIV>> As you find, only when the NSD is deployed on the TA, the performance</DIV>
<DIV>> (e.g., QPS) can be promoted significantly. In other words, even when the</DIV>
<DIV>> SA uses the NSD, the QPS is just similar with the case when all the</DIV>
<DIV>> servers use BIND. This confused me, so why the results are totally</DIV>
<DIV>> different when TA uses NSD and SA uses NSD.</DIV>
<DIV>> </DIV>
<DIV>> In my mind, the results should be similar in these two cases.</DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> This same results appear when the recursive server uses UNBOUND. And the</DIV>
<DIV>> corresponding results is listed as following:</DIV>
<DIV>> </DIV>
<DIV>> Case1: Recursive(unbound)---Root (bind)---TA*(bind)*---*SA (bind)*</DIV>
<DIV>> </DIV>
<DIV>> 254.047935 qps</DIV>
<DIV>> </DIV>
<DIV>> Case2: Recursive(unbound)---Root (bind)---TA*(bind)*---*SA (nsd)*</DIV>
<DIV>> </DIV>
<DIV>> 293.855172 qps</DIV>
<DIV>> </DIV>
<DIV>> Case3: Recursive(unbound)---Root (bind)---TA*(nsd)*---*SA (bind)*</DIV>
<DIV>> </DIV>
<DIV>> 14968.491326 qps</DIV>
<DIV>> </DIV>
<DIV>> Case4: Recursive(unbound)---Root (bind)---TA*(nsd)*---*SA (nsd)*</DIV>
<DIV>> </DIV>
<DIV>> 15071.630940 qps</DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> Thank you very much for your attention!</DIV>
<DIV>> </DIV>
<DIV>> Zhiwei Yan</DIV>
<DIV>> </DIV>
<DIV>> 2011-09-16</DIV>
<DIV>> ------------------------------------------------------------------------</DIV>
<DIV>> yzw_iplab</DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> </DIV>
<DIV>> _______________________________________________</DIV>
<DIV>> nsd-users mailing list</DIV>
<DIV>> nsd-users@NLnetLabs.nl</DIV>
<DIV>> http://open.nlnetlabs.nl/mailman/listinfo/nsd-users</DIV>
<DIV></DIV>
<DIV>-----BEGIN PGP SIGNATURE-----</DIV>
<DIV>Version: GnuPG v1.4.11 (GNU/Linux)</DIV>
<DIV>Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/</DIV>
<DIV></DIV>
<DIV>iQIcBAEBAgAGBQJOcweTAAoJEJ9vHC1+BF+NioMP/0t2tCOIQ9e4sWCBWqAGV+Tq</DIV>
<DIV>Z3FaIPF31TYh8JnKQFjDGH5oej7y2y05iOe6nr9dUkXKydjntAhejZnVdK5Om9ez</DIV>
<DIV>ImFwQSYuWIDvxE3IV4Ke8X7i73bJwLS23p9caBwefBLWI6wOxB0q8BSvaMMDdkGK</DIV>
<DIV>5+nzp7IKMm4oW2IUm31eC96Q96/RYke+mJdR4hBE+pTwZRt3rPJ1ShkaahJi8vWv</DIV>
<DIV>m8MZSsr2jw6U/ZGZ+HaOdpCfyNU1RGOO0Sf99SATXU7nUXCI0uC8GLItnYFU0A3M</DIV>
<DIV>u4py43SHnipSL9VNVLs7aNKMQ0SAf/FgYerSwZ33Np+zCEgWfBZS2HVYO/KCBr1t</DIV>
<DIV>sBCsi5pXjMG596wWg8y1p93Ft1G6CTMv9d0aUgzIYaNXNOaRiPiraDENpoMAPEKQ</DIV>
<DIV>3bITbtqgk8LC7JGb/6I67yaWwRXESREWCKQg0aA+GSaypZf6T0as9kYwEtgFjYrX</DIV>
<DIV>+eRPENIFfSQWyCB3l7R5zVQeEPyUv1W2k/yCvlCQJRT6Y5K1H/z06/t1oB15B7MD</DIV>
<DIV>45Hl2vmmFqDfhW5up3ieVRSi9nyh4EdNcZu95QtbKbg25CyOKfXxP7msVcr3vs8j</DIV>
<DIV>9feLPgyow4iqPRBE1FHktzG3nTrmJVtAhasbiGHXESIx7YDyzQ/4Q9kwcB/kKDss</DIV>
<DIV>cToaSglqTWpTeP7I358J</DIV>
<DIV>=/seK</DIV>
<DIV>-----END PGP SIGNATURE-----</DIV>
<DIV>_______________________________________________</DIV>
<DIV>nsd-users mailing list</DIV>
<DIV>nsd-users@NLnetLabs.nl</DIV>
<DIV>http://open.nlnetlabs.nl/mailman/listinfo/nsd-users</DIV></FONT></DIV></BODY></HTML>