COOK: does this mean that there is some way to get the cisco to behave
like the cascade switch I asked about?? If not what in general does it mean?
No, this means that we should be able to squeeze every
last drop of bandwidth out of the 7500, such that 4x+ or so
worth of OC3s can work in the same chassis without
Finally you said if the 7500 doesn't perform well then you might have to
do the redesign I asked about.... meaning going to an atm switched
backbone with routers at the periphery of the backbone??
Over my dead body. The nicer bits of Curtis's last message
very clearly and correctly indicated the drawbacks in using
ATM or any other L2 protocol. Perhaps he or someone else
can put it into slightly less technical terms.
No, what a disaster would mean is replacing backbone 7500s
with 7000s and doing lots of thinking about really ugly
kludges to deal with scale beyond the bandwidth limits of
the 7000 series backplane and lots of screaming at Cisco to
make the 7500 work at least as well as the 7000+SSP.
(We have the latter committment from them already, btw.)
Many possibilities exist that don't entail going with ATM,
or even necessarily with some non-Cisco switching engine, or
some L2 protocol sandwiched between the physical lines and
IP, all of which would work in principle.
As I think I have said earlier, the value of Cisco is much
more in its software than in its hardware; I like Cisco
because I know that IOS can be made to work the way I need
it to, which is more than I can say for any vendor of any
other platform. Cisco does weird things with respect to
hardware that sometimes gets annoying or frustrating, but as
Melinda Shore's .signature (after Hippocrates) reminds the
world: "software longa, hardware brevis".
P.S.: Actually, the quote from Hippocrates's _Aphorisms_ is
pretty apt to this discussion:
"[Of medicine] The life so short, the art so long
to learn, opportunity fleeting, experience treacherous,
Maybe Dave Mills should consider storing this with
his jaguar.txt file.