QOS or more bandwidth

Sean M. Doran smd at clock.org
Tue May 29 18:13:42 UTC 2001

Prabhu Kavi writes:

| Someone asked earlier in this thread if it was cheaper to add
| capacity or pay for the bright engineers to make TE or QoS work.
| For large carriers, the right answer is often to pay for the
| bright engineers.

Admittedly I have strong biases, but the engineers that I think
are bright will tell large carriers that the right answer is to
spend money on more capacity.

They are also the people _least_ likely to be impressed by
money as the currency "to pay for" them.  Being taken seriously
despite having very non-bell-head ideas, and having an environment
in which traditional thinking does not win every argument, is
usually much more important.

I've done some non-TE work for a couple of networks, and
the only bandwidth-constraint they have run into on the
technical front has been switching capacity of routers.
Yeah, you sometimes have to fight to "remove" fun features
like 1+1 protection and monitoring, but this can be done
even in the most traditional of telcos, as demonstrated by
various people over the years.

| Of course, the same amount of bandwidth they had then 
| would now cost much less, and be considered a small network, 
| and the results today could well be different.

In my necks of the woods the retail pricing of international
TDM (PDH, SDH, WDM) capacity eroded something like this:

	1993 - T1 capacity, 9-12 cents/channel-mile
	1994 - T3 capacity, 6-9 cents/channel-mile
        1995 - OC3 capacity, 3-4 cents/channel-mile
	1997 - (intl) E3/T3, midpoint, $4M (US) we don't count in channel-miles
	2001 - $4M (US) gets you a multi-city optical subnet at 2.5Gbps

What "we" believed in 1995-1997 about ATM cell tax and the like
is no longer valid.  Neither is what "they" believed about traffic management.

| 100 Nagog Park                  WWW:    www.tenornetworks.com

Wow I misread this street address several times...  lysdexia, I guess.


More information about the NANOG mailing list