Cell Relay Archive

Cell Relay Retreat>List Archive>month:2000-Aug> msg00035



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Re: CR: Re: ATM vs. Frame Relay

  • From: Afroz Lateef <afroz.lateef@fnc.fujitsu.com>
  • Date: Wed, 09 Aug 2000 12:03:11 -0500
  • CC: cell-relay <cell-relay@cell.onecall.net>
  • Organization: Fujitsu Network Corporation

Thomas,

For the SPVC re-routing, when a link fails, is mentioned in Annex C of PNNI1.0,
in the third paragraph.
As such, when the call is attempted to be re-established, then since the original
link or switch could have failed, the new set-up message is cranked back and
alternate route used.

Reg: VNN, You are right.  I confused it with VNNI (Virtual Network-Network
Interface) and not Ascend's VNN.

Sincerely

Afroz Lateef

Thomas Frieling wrote:

> Afroz Lateef schrieb:
> >
> > Thomas and Charles,
> >
> > PNNI does have re-routing facilities (of course, for  SPVCs and SPVPs only).
> > Crankback and re-routing in the PNNI spec, I guess talk about this.
> > VNN is not Lucent's proprietary, it is part of the specs. (VNNI and VUNI).
> >
> > Afroz Lateef
> > Fujitsu Network Communications
> >
>
> Alfroz,
>
> where is this ability quoted? I have searched the ATM-Forum spec
> (PNNI 1.0) and the SPVC Addendum and there are no facilities
> defined to dynamically re-route broken connections. Crankback is
> a mechanism that is used when the route calculated by the source
> switch is broken at setup time! Crankback makes sure that a
> viable route can be found in that case. Once the connection is
> established PNNI has finished its job.
> Connection re-routing is under way in the ATM-Forum study group
> Control Signalling and currently in the status "Work in Progress
> -- to be done".
> And VNN is definitely proprietary to Ascend (now part of Lucent).
> This can for example be seen in
> http://www.ascend.com.au/2373.html where you find this: "...
> Virtual Network Navigator (VNN), Ascend's enhanced version of the
> industry-standard OSPF algorithm, ...".
>
> Regards
>
> Thomas Frieling