The Routing Over Large Clouds Mailing List Archive by date

Cell Relay Retreat>List Archive>month:1994-Sep> msg00005



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

ROLC and IPng

  • From: Dave Katz <dkatz@cisco.com>
  • Date: Wed, 28 Sep 1994 14:11:56 -0700
  • Cc: rolc@maelstrom.timeplex.com

There is a bit of bickering, er, thoughtful argument going on about whether
hosts should be "subnet knowledgable" or not.  Effectively, NHRP tosses
subnet knowledge to the wind anyhow, so in some sense it doesn't really
matter how it settles--for large clouds, NHRP makes the subnet model
(from the standpoint of hosts, anyhow) irrelevant.

So I would envision NHRP being essentially the same for IPng as IPv4.

   From: shur@arch4.ho.att.com
   Date: Wed, 28 Sep 94 13:12:15 EDT

   How does the problem of ROLC carry over into the IPng environment?
   Will IPng hosts operate like IPv4 hosts and transmit either to
   destinations with the same network prefix, or to a first-hop router for
   destinations with a different network prefix? Or will they employ a
   different algorithm which may enable them to reach non-local
   destinations on the same large cloud without needing to be forwarded by
   intermediate routers? If the latter, how does the aforesaid algorithm relate to NHRP?

   Thanks,
   David Shur.
   (d.shur@att.com)