Another reason for the DLCI limitation is the number of Interface Descriptor
Blocks (e.g. subinterfaces, loopbacks, dialer) available per platform - see:
http://www.cisco.com/warp/customer/63/idb_limit.html
When the IDG was first published, this limitation had not been exploited.
As time went on (i.e. as TAC got more calls), Cisco decided to publish the
limitation, and it finally made it to the tech writers somehow. Probably
via a tech editor who was up on his/her IOS information, or maybe even
someone from the TAC was on the review list.
Of course, if you want to enter more than 300 frame-relay map statements,
that would be your prerogative ;-)
-e-
----- Original Message -----
From: Chuck Larrieu
To:
Sent: Monday, July 09, 2001 12:31 AM
Subject: RE: Capacity guidelines [7:11174]
> I gather you have CD's from a ways back.
>
> BTW, did you notice the typo? Low end boxes ( 4,500 and below ) was that
> dollars? or model numbers?
>
> If I get the drift of the quote, plus some of the other stuff in the IDG,
> static routing ( i.e. no routing protocols enables ) permit more DLCI's on
> the same interface because of low routing overhead.
>
> BTW, ever looked at the output of a debug frame packet? that along seems
> awfully chatty.
>
> thanks for the reminder. there really is a lot of good stuff on CCO when
you
> can find it.
>
> Chuck
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
> [EMAIL PROTECTED]
> Sent: Sunday, July 08, 2001 11:51 PM
> To: [EMAIL PROTECTED]
> Subject: RE: Capacity guidelines [7:11174]
>
>
> The Internetwork Design Guide has increased its guidelines over the years
:
> -)
> >From the IDG - January 1994...
> "In general,Frame Relay designs should feature a maximum of 10 to 50 DLCIs
> per interface in a given internetwork. The specific number depends on
> several factors that should be considered together:"
>
> >From the IDG - July 2001...
> "How many DLCIs can be configured per serial port? It varies depending on
> the traffic level. You can use all of them (about 1,000), but in common
> use, 200-300 is a typical maximum. If you broadcast on the DLCIs, 30-50 is
> more realistic due to CPU overhead in generating broadcasts. Specific
> guidelines are difficult because overhead varies by configuration.
However,
> on low-end boxes (4,500 and below), the architecture is bound by the
> available I/O memory. The specific number depends on several factors that
> should be considered together: "
>
> Both editions then go on to state exactly the same factors (protocols,
> level of broadcast traffic, line speed etc), in the same words. Including
> a rather dubious statement that is either wrong, or I'm just
> mis-interpreting it. "If static routing is implemented, you can use a
> larger number of DLCIs per line, because a larger number of DLCIs reduces
> the level of broadcasting". Sounds like a dodgy reason to me.
>
> It's actually not a bad read.
>
http://www.cisco.com/univercd/cc/td/doc/cisintwk/idg4/nd2009.htm#xtocid13460
> 13
>
> JMcL
>
> ---------------------- Forwarded by Jenny Mcleod/NSO/CSDA on 09/07/2001
> 04:25 pm ---------------------------
>
>
> "Chuck Larrieu" @groupstudy.com on 09/07/2001 08:47:25
> am
>
> Please respond to "Chuck Larrieu"
>
> Sent by: [EMAIL PROTECTED]
>
>
>
> To: [EMAIL PROTECTED]
> cc:
>
>
> Subject: RE: Capacity guidelines [7:11174]
>
>
> let's see ( putting on my thinking cap )
>
> the DLCI field of the frame relay header is 10 bits, meaning a maximum of
> 1023 DLCI numbers.
>
> did not find anything on the frame relay forum site www.frforum.com, but
> according to Darren Spohn's Data Network Design ( out of print but a third
> edition is being written ):
>
> DLCI 0 = in channel signaling ( lmi-type ansi )
> DLCI 1-15 are reserved
> DLCI 1008-1022 are reserved ( I always thought it was 1005-1022. wonder
> where I got that from )
> DLCI 1023 = in channel layer management ( lmi-type cisco )
>
> that leaves 16-1007 as possible DLCI's for production - or 992 total.
>
> interesting.
>
> Chuck
>
> -----Original Message-----
> From: Mike Mandulak [mailto:[EMAIL PROTECTED]]
> Sent: Sunday, July 08, 2001 12:38 PM
> To: Chuck Larrieu
> Subject: Re: Capacity guidelines [7:11174]
>
>
> As I recall the theoretical limit on the number of DLCIs is 999. This from
> a
> Mark Miller book. The book is at work so I can't check it right now. Not
> sure of vendor specific limits either.
>
>
> ----- Original Message -----
> From: "Chuck Larrieu"
> To:
> Sent: Sunday, July 08, 2001 3:00 PM
> Subject: RE: Capacity guidelines [7:11174]
>
>
> > I'm curious about this statement. Can you clarify?
> >
> > I know I've read on CCO about limits to numbers of subinterfaces (
> actually
> > to total numbers of interfaces in general ), due to constraints of the
> > router architecture.
> >
> > I don't believe I have come across any reference to limits on the number
> of
> > DLCI's supported, although it wouldn't surprise me to learn that there
> are
> > limits for whatever reason - hardware, memory, or architectural
> > considerations. Or Telco side limitations.
> >
> > Chuck
> >
> > -----Original Message-----
> > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> > Sent: Sunday, July 08, 2001 11:25 AM
> > To: [EMAIL PROTECTED]
> > Subject: RE: Capacity guidelines [7:11174]
> >
> >
> > Frame relay serial circuits commonly support 200-330 dlci's as a
maximum.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=11421&t=11174
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
