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

Fwd: Testing @ IETF Meeting



Dear IESG Members:

I am forwarding a message from Brett Thorson, a member of the Foretec technical team, that details some of his concerns about interoperability testing before, during, and after an IETF meeting. While all of the issues that he raises may not apply in all cases, they do reflect a somewhat different perspective on the situation, and you may want to consider them in your deliberations.

Regards,

Barbara
----------------------------------------------------------------
Envelope-to: bfuller@foretec.com
From: Brett Thorson <bthorson@foretec.com>
To: "Barbara B. Fuller" <bfuller@foretec.com>
Subject: Testing @ IETF Meeting
Date: Thu, 11 Sep 2003 14:20:22 -0400
User-Agent: KMail/1.5

If the IESG decides to approve of the testing, they should be aware of the
added difficulties & cost that it adds to the host and/or Foretec.

From some previous e-mails it reads that some of these testers were expecting
the network to be up and running when they got there (which would be before
the meeting). Trying to setup a network with "experiments" going on,
increases the level of stress while trying to get this network up and
running. On the back end, leaving the network run longer for testers is just
that many more days added onto a really long week (for the NOC Crew).


One of the considerations should also be the cost of this event.  If we are
going to sanction testing before or after the meeting, some level of network
infrastructure will need to be setup before or remain after the meeting.

Even now, we have a hard time getting into these rooms before the meeting. The
hotel would rather sell them to a customer, rather than letting us in for
free before we have officially purchased the room. Also with the hotels now
adding more wiring to the buildings, and the IETF using that wiring, we also
incur the additional costs of using their infrastructure for these days. In
some cases (Such as Minneapolis) we have received gracious considerations
from both Onvoy and U of M in providing Internet circuits. Right now, they
are donating these to the IETF. Would we then need to offer some sort of fee
to them for using their networks beyond the official IETF meeting?


If we get into a hotel that doesn't mind us being there early, and lets us
leave the equipment there for as long as we want, or come in as early as we
want, then we still have the issue of people to set it up.  (Hotel rooms,
meals, missed work, etc.)

All costs aside, it also encompasses a face to face issue.  The relationship
that any host or secretariat has with the hotel is important.  If you get a
good team at the hotel (and treat them right) the meeting is A LOT easier.
If these groups were to offer that they would show up at the hotel early, set
up their own equipment, I would also fear that these testers might not give
the hotel a good impression of the IETF.  That would make our job (coming in
the week after, etc.) that much harder.  The telecom guys especially, you
want to keep them happy, and starting off with a positive relationship is
VERY important.

Also take into consideration that the network quality expected at an IETF
meeting, is VERY different than the network expected at an Interop meeting.
People come to the IETF to work and attend meetings. I don't know a lot about
the Interop testing grounds, but I do know that they have a hot-stage event.
IETF usually does not.  Most of our equipment (when Foretec has hosted the
meeting) is loaned and borrowed, and we usually don't get it until close to
meeting time.  Making this hardware plug and play, with IETF expectations is
a different scenario then getting hardware early, hotstaging it, and finding
bugs and kinks early.

The possibility of a testing event is not out of our grasp.  All the
challenges listed here have solutions.  However, it doesn't appear that any
of the solutions are free.  Nothing to say that it can't be done.  I do think
having an open discussions about the costs & benefits of this event is worth
while.

Respectfully submitted,
Brett M. Thorson
------------------------------------------------
Barbara B. Fuller
Acting Executive Director
Internet Engineering Task Force

Natick, Massachusetts Office:

Phone: +1-508-650-4020
Fax:     +1-508-650-4639

Reston, Virginia Office:

Phone: +1-703-620-9053
Fax:     +1-703-620-9071