Dave Sill <[EMAIL PROTECTED]> writes on 4 August 2000 at 09:37:29 -0400
 > "David Dyer-Bennet" <[EMAIL PROTECTED]> wrote:
 > 
 > >Dave Sill <[EMAIL PROTECTED]> writes on 2 August 2000 at 10:14:56 -0400
 > > > 
 > > >   http://www.kyoto.wide.ad.jp/mta/eval1/eindex.html
 > >
 > >His methodology looks reasonably sound, now that I can read the
 > >description of it.  And he seems entirely aware of the shortcomings,
 > >which leads me to trust his judgement on other matters as well.
 > >
 > >Looks like qmail took 20 seconds and sendmail took 1750 seconds to
 > >deliver his test load.  Not surprising!  (uncached case)
 > 
 > I don't see where you got 20 seconds. Here's the results in tabular
 > form--numbers are all APPROXIMATE since I'm reading them from the
 > graphs (the individual results by implementation):
 > 
 >              Eval 1          Eval 2          Eval 3
 > MTA       time    dns     time    dns     time    dns
 > qmail      155   1250      127   1230      127   1235
 > Postfix    184   1375      168   1290      161   1330
 > exim       645    475      161    450      157    451
 > SMTPfeed   215    610      160    442      157    461
 > zmailer   1530   1675      357   1260      360   1300

I read the time on eval 1 for qmail as 20 seconds.  Well, maybe 22.
There's a very sharp bend in both DNS and SMTP curves at that point,
and only completely trivial activity after that.  I do see that the
DNS answer curve is measurable separated from the DNS request curve;
but the SMTP lines don't appear to change after that, so whatever DNS
is doing, delivery has completed.

 > >Also note that in the cached case postfix appears to beat qmail at
 > >delivering all the mail, at least on one graph.
 > 
 > I don't see that.

Well, maybe not, the SMTP fin line is separated a bit from the syn
line which the computed line is based on.

 > >However, did people notice that sendmail actually did *fewer* DNS
 > >queries?  I had understood that for total bandwidth use, qmail won
 > >over sendmail partly for doing less DNS traffic, but this doesn't seem
 > >to be the case in this study.
 > 
 > Yeah, that suprised me, too. Exim wins the prize for DNS frugality,
 > though.
 > 
 > >(postfix took 30 seconds, exim 500, zmailer I can't tell.  Am I
 > >reading the graphs wrong?
 > 
 > Where are you seeing these numbers?

Eval 1, the individual graphs mostly.  I'm using the point where the
SMTP fin count maxes as the terminal point, even though some DNS
activity occurs after that with some mailers.

But I don't see why I was confused about zmailer now (other than the
trailing DNS activity), seems to finish at about 190.

 > >Zmailer shows increasing count of DNS
 > >queries off to the end of the map, but no increase in SMTP syn or
 > >fin.  Now I'm confused.)
 > 
 > Me too, because I just don't see that. Which graph(s) are you looking
 > at?

http://www.kyoto.wide.ad.jp/mta/eval1/perf1-zmailer.gif  (evaluation
1, zmailer).  The SMTP syn count has peaked a bit under 200 seconds,
the SMTP fin count shortly thereafter.  The DNS query and response
count are at about 1275 then.  By 1400 seconds, the DNS query and
response count are up to about 1550.
-- 
Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b 
David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]

Reply via email to