On Feb 19, 2008, at 12:36 PM, Randall Atkinson wrote:
Earlier, Dow Street noted: % That's correct - this would be a new per-packet processing cost, % probably implemented in hardware. I'll leave it to the equipment % designers to comment on the implementation cost / complexity. This would be a major increase in equipment costs to ISPs, as near as I can tell, because changes to per-packet-processing that require new hardware are not low cost. The burden on smaller access routers, which often still use CPU-based forwarding would also be an issue (software is easier to update, but the increase in number of CPU cycles per packet is an issue for deployed routers).
Thanks, Ran. I've mostly been thinking about fundamental / long-term changes to the architecture, so I'm expecting some cost. I'm not necessarily convinced this specific mechanism would be worth the cost - just beginning to think through it. Any thoughts on how this mechanism would compare to other changes to router hardware? (MPLS, IPv6, LPM, ACLs, encryption, etc)
As Elwyn said, I think this could be incrementally deployable (since it is only used as a hint).
R, Dow -- to unsubscribe send a message to rrg-request@psg.com with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg