[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Comments on draft-ietf-multi6-multihoming-requirements-00





> I wish the same could be said about IPv6 and routers. :(

Just as comment.  Actually doing this in hardware.  There is no extra cost
except for the pipeline LA to Register has 12 more bits if you design a good
RIB and FIB on your cpus.  Whether you use ASIC or NP and whether you use
SRAM or CAM to get the dst addr.  The cost now that makes these 12 bits
insignificant is the tunnels we must now use and looking into the
tcp/udp/sctp data portion of the packet.  Thus far the next hdr is not an
issue unless your in the hop-by-hop path or a Mobile Home Agent (in case of
finding the dst option).  And if we use the flow label e2e the lookups for
all processing will get better.

/jim

>