The Routing Over Large Clouds Mailing List Archive by date

[Curtis Villamizar <curtis@ans.net>]

In message <199602220456.XAA02493@shovel.nexen.com>, James Luciani writes:
> Folks...
> 
>   I just submitted the following draft to deal with the server 
> synchronization issues.  I tried to use ideas from 
> well known/understood/tested technologies such as OSPF and PNNI which 
> hopefully also translates to maximum code reuse and a pre-existing slew 
> of folks who understand the general concepts.  Thanks to all the folks 
> who had the opportunity to give feedback!  I apologize for the size but
> I tried to be as detailed as possible while still solving the problem in
> a general enough way so that we don't end up with a bunch of point solutions
> for each protocol (i.e, NHRP, MARS, ATMARP, etc.).  Even still, as you can se
> e, 
> the state machines are pretty simple and the excitations are 
> reasonably straightforward.
> 
>   I probably won't be able to cut another NHRP draft and have it in for
> the I-D cut-off but it is possible since I tried to do updates as suggestions
> came in.  However, if humanly possible, I will get a copy out onto the mailin
> g list
> before LA.
> 
> Regards,
> -- Jim Luciani
> __________________________________________________________________________
> James V. Luciani    Ascom Nexion                    voice: +1 508 266-3450
> luciani@nexen.com   289 Great Rd., Acton MA 01720   FAX: +1 508 266-2300


This seems like a clone of the OSPF support for broadcast or NBMA.
The problem with this as size of the LIS grows is that traffic and
server performance hot spot created by having everything redistributed
by the DR.  (In other words, I agree with what Joel said).

The rumblings at Dallas were toward go with something like OSPF
flooding, as a special case of the distributed database algorithms in
the RFC1577++ draft where the number of adjacencies to inform was
equal to the number of adjacencies (rather than allowing them to be
less than the number of actual adjacencies and creating the potential
for incomplete flooding).  This would involve modeling the mesh of
servers very much like a mesh of OSPF ptp virtual links.  There would
be no advertisement of adjacencies and SPF calculation.  The
information being flooded would be analogous to OSPF AS external (ASE)
routing information and simply collected rather than used in
conjunction with the results of an SPF.

Curtis