ietf
[Top] [All Lists]

Re: (ietf54-noc 1802) Re: why we had wireless problems at IETF

2002-07-17 22:29:54
At 10:58 AM 7/18/02, ggm(_at_)apnic(_dot_)net wrote:

> >the accompaning issues is more than 60-100 clients per ap (and ~200=death)
> >really results in reduced performance as well, particulallry if most of
> >them are active so more ap's can result in better localized performance,
> >assuming you get a handle on the rf issue.
>
> maybe at IETF55, should we invite 802.11b experts to measure behaviors,
>       expreiment with basestation placement, and such?
>

I very much agree. I think we can eat our own dogfood here!

At the Minneapolis meeting, the wireless network worked quite well. Agere / Orinoco provided the equipment and engineers who knew how to properly set it up. I saw few if any problems in Minneapolis. Having spent the time to become a certified reseller, there are lots of things that can affect sites, and ways to deal with large meetings. I do not know if the Yokohama network was engineered properly or not. I did see some email at the start of the week which indicated the channels were not well chosen, though it appears that has been remedied.

It might make sense to just spend the money to properly design these networks, even if they're for short duration use.

-----------------------------------------------------------------
Daniel Senie                                        dts(_at_)senie(_dot_)com
Amaranth Networks Inc.                    http://www.amaranth.com