[e2e] QoS vs Bandwidth Overprovisioning

[Vernon Schryver]

> From: Rajesh Talpade <rrt at research.telcordia.com>

> ...
> While it is indeed true that bandwidth in general is available in plenty
> at the current moment, there may still be a place for QoS. QoS does not
> need to imply the "multiple handshake, back-slapping, etc" protocols
> that network elements need to go thru to "reserve resources." It can be
> as simple as provisioning network links to support a few traffic classes.
> ...

Has anyone (esp. major carriers) tried the simple, obvious approach of
honoring the old-style IP TOS bits?
Yes, I remember that "everyone" says that without accounting to prevent
cheating, "all users" would soon mark all of their packets low-delay,
but does that actually happen?  In the absense of cheating, and as
long as voice-over-IP and other latency sensitive and very loss
sensitive applications are a minor fraction of the traffic, wouldn't
old-style IP TOS do fine?  Has any large ISP tried honoring the old
TOS bits so end users could see what happens?

Concerning secure tunnels to work thanks to cable modems, T3's and so
forth, my personal experience is somewhat different than reported here
recently.  I've done a little ssh tunneling to distant @Home customers on
the West Coast and others on the East, and not been overjoied.  Typical
latencies up to 2 seconds with simultaneous packet losses up to 30% for
non-bulk traffic tend to discourage even the most primitive interactive
applications such as dumb terminal stuff.  Things like X-windows through
such tunnels must be avoided except when absolutely necessary when you
will tolerate literally several minutes per interaction.  I've also moved
a few 10's of GBytes through such pipes (necessary test data for something
that didn't quite fit the FedEx-CDROM mode), using patience and large
TCP windows.

Maybe things are not as bad for people using the same retail ISP as
their local employer or university.  However, I don't understand what
Tier 1 IPSs and transatlantic or transcontinental dark fiber have to
do with such situations.  Maybe MAE-West and other exchange points
are not as bad now as they were months ago, but somehow, after all
these years, I doubt it.  I also don't understand much of interest
for fancier QoS than old-style IP TOS in such local situations.


Vernon Schryver    vjs at rhyolite.com