[c-nsp] Your opinions on router throughput

[Ted Mittelstaedt]

Hi All,

  It's OK guys, I asked a somewhat open-ended question because I wanted
the discussion.

  I already know what both VXR's with the NPE300's are doing in their
existing
circumstances - show proc gives me that.  Right now they are running at 10%
CPU utilization with cef turned on and 2 BGP views.  It is useful to know
that
others are seeing the same utiliztion in "normal" circumstances.

  It is also almost even more useful to know how the devices will react in
"abnormal" circumstances. :-)

Ted


----- Original Message ----- 
From: "Tony Varriale" <tvarriale at comcast.net>
To: <cisco-nsp at puck.nether.net>
Sent: Sunday, October 29, 2006 2:53 PM
Subject: Re: [c-nsp] Your opinions on router throughput


> There were a lot of opinions that were expressed that were not necessarily
> on-topic and somewhat in left field.  But, you chose to focus on my
> posts...once again.
>
> I'm more than willing to take this off-list if you feel some sort of
> attraction to my posts.  This way we can keep this on-topic for the
original
> poster.
>
> tv
> ----- Original Message ----- 
> From: "Gert Doering" <gert at greenie.muc.de>
> To: "Tony Varriale" <tvarriale at comcast.net>
> Cc: <cisco-nsp at puck.nether.net>
> Sent: Sunday, October 29, 2006 3:29 PM
> Subject: Re: [c-nsp] Your opinions on router throughput
>
>
> > Hi,
> >
> > On Sun, Oct 29, 2006 at 12:01:53PM -0600, Tony Varriale wrote:
> >> Have you ever seen a non-hardware based Cisco products do MLPPP?
> >
> > Indeed...
> >
> > Cisco-M-XIII>sh ppp mu
> >
> > Multilink200, bundle name XXXX
> >  Bundle up for 6w2d
> >  570 lost fragments, 18471806 reordered, 0 unassigned
> >  57 discarded, 0 lost received, 1/255 load
> >  0x18CE3E received sequence, 0x7A8F55 sent sequence
> >  Member links: 2 active, 0 inactive (max not set, min not set)
> >    Se3/7:0, since 2w5d, last rcvd seq 18CE3C
> >    Se3/3:0, since 3d12h, last rcvd seq 18CE3D
> >
> > Cisco-M-XIII>sh ver
> > ...
> > cisco 7206VXR (NPE225) processor (revision A) with 114688K/16384K
> > ...
> >
> > Of course this box is not very heavily loaded - the MLPPP bundles
> > are only 2x E1 each.  Add to that a few more E1s, one channelized E3,
> > and two T3s and two E3s, with usually about 40-50 Mbit/s peak passing
> > through the box.   CPU load usually well below 35% (and that's a
NPE-225,
> > not a NPE-300).
> >
> > 12.2(18)S10, CEF, IPv4+IPv6, 51 weeks uptime, lots of happy packets.
> >
> >> Or some good ol' voice?
> >
> > I haven't used a Cisco box for voice termination yet.
> >
> > OTOH our core network is carrying our own voice traffic (between Snom
> > phones and an Asterisk PABX) via STM-1s connecting NPE-400s and NPE-G1s
> > - but a STM-1 with imix traffic pattern won't saturate a NPE-400, so
> > we haven't run into any router-induced issues with the VoIP stuff yet.
> >
> >> Just FYI, I'm not asking this in condescending manner...I'm
> >> asking becuase you didn't offer this up as an example.
> >
> > Well - the original poster *did* *not* *ask* for such.
> >
> > It would be so helpful if people wouldn't just stray off from the
> > *questions* people ask just to tell about their grandchildren...
> >
> >> Oh...could you point me to the CEF sucks info thread?  I must of missed
> >> that
> >> one. :)
> >
> > Just figure out how to use the list archives.  There are enough postings
> > where people recommend to turn off CEF, or where people state that they
> > are not running CEF because this-or-that problem.  Or other people
> > claiming
> > that "CEF" is an acronym for "customer enragement feature".
> >
> > Ceterum censeo: if you're doing something extraordinary to your NPE-300
> > to reach 70% CPU load with 10 Mbit/s. of load - congratulations to you,
> > this is an achievement.  But please don't offer this as meaningful
advice
> > to someone asking for "how will a NPE-300 behave under *normal*
> > circumstances".
> >
> > gert
> >
> > -- 
> > USENET is *not* the non-clickable part of WWW!
> >
> > //www.muc.de/~gert/
> > Gert Doering - Munich, Germany
> > gert at greenie.muc.de
> > fax: +49-89-35655025
> > gert at net.informatik.tu-muenchen.de
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>