From: M e (iosluver@gmail.com)
Date: Thu May 22 2008 - 15:03:31 ART
Gents,
Many thanks for the input on this it's highly appreciated. I was leaning
towards the points you made regarding the improvement in CPU power which
network devices have benefited from in many way.
I raised the question of limits on number of OSPF areas in a situation where
high end devices were not available. For instance a solution was required on
a network of 38000 sites as sited in the doc at (
http://www.netcraftsmen.net/welcher/papers/ospf1.html) & the luxury of
modern routers with high end CPUs wasn't on the table for cost or any other
fictitious reason. That's flimsy because anyone with 30k sites should be
able to afford an MLS but crazier things have happened & donkeys have flown
before ask folks in Verona Anyways work with me here.. If the choice of
using stub areas was an option & that was taken. For the end result to be
objective to be reareasonable, there would be quite a number of areas which
would be upwards of a 100+ (conservative estimate assuming the idea was to
reduce LSDB size) . Put another way, what's the largest number of areas any
one has seen in production before?
Cheers
On Thu, May 22, 2008 at 12:01 PM, Salau, Yemi <yemi.salau@siemens.com>
wrote:
> That was just a "floor-walking" report of what was in existence then
> (1998); I'm sure that table would have been way stretched in the 21st
> Century.
>
> I stumbled across Dr. Peter J. Welcher's take on this subject, under the
> OSPF Design Cisco Rule of Thumb:
> http://www.netcraftsmen.net/welcher/papers/ospf1.htm
>
> Many Thanks
>
> Yemi Salau
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> dara tomar
> Sent: Thursday, May 22, 2008 11:26 AM
> To: sheherezada@gmail.com
> Cc: iosluver@gmail.com; ccielab@groupstudy.com
> Subject: Re: Limit on OSPF Areas
>
> *A good oldie question,
>
>
> Beside what Mihai said all is dependent on the CPU-processing power and
> the
> memory consumption of a device.
>
> Here is what the RF 2329 as per the year 1998 has to say.
>
> http://www3.tools.ietf.org/html/rfc2329#page-7
>
> And with time we are having more powerful devices in market, so no real
> concerns...
>
> Regards,
> Dara*
>
>
> On Thu, May 22, 2008 at 1:46 PM, <sheherezada@gmail.com> wrote:
>
> > I don't think this is the case with the current processing power on
> > the routers. I've seen 100 routers in a single area. Actually,
> > single area OSPF is better especially in SP design, where you carry
> > all bulk routes in BGP and avoid TE complications.
> >
> > Just my opinion.
> >
> > Mihai
> >
> > On Wed, May 21, 2008 at 9:55 PM, <iosluver@gmail.com> wrote:
> > > Hi,
> > >
> > > I have read & heard about the need to limit the number of routers in
> an
> > OSPF
> > > because of the performance implications on participating devices. Is
> > there a
> > > limit on the number of OSPF Areas?
> > >
> > > Thanks in advance.
> > >
> > >
> > >
> _______________________________________________________________________
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> >
> >
> >
> _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Mon Jun 02 2008 - 06:59:18 ART