[ietf86-tech] Fwd: WAPs for Orlando

Fri Feb 8 10:44:12 PST 2013

On Thu, Feb 7, 2013 at 1:04 PM, Chris Elliott <chelliot at pobox.com> wrote:

> On Feb 7, 2013, at 12:38 PM, joel jaeggli <joelja at bogus.com> wrote:
>
> > On 2/7/13 9:33 AM, Chris Elliott wrote:
> >> Was in a meeting...out now...
> >>
> >> I'm assuming the 1142s have dual radios. If not, no way.
> > if it has one it's an 1141
> >> The real performance difference, given both have dual radios, is that
> the 1252s have support for 802.11n with three antennas per radio and MIMO
> support, in addition to .11a,b,g support, which only gives you two antennas
> with diversity. This support helps all clients, not just .11n clients,
> including 2.4Ghz clients where we don't enable .11n--we still get benefits
> from the better radio for both transmission and reception.
> > 1142 is 2 x 3 mimo
>
> My bad. The radios are the same (or maybe a bit better in the 1142s as
> it's newer), --conflated the earlier 1132 and 1242 APs specs in my brain.
> Same on CPU/memory and uplink (gig). The only advantage for the 1252s would
> be the better antennas, which may not actually be beneficial in this
> environment. And i verified that the 1142s can do .11n on both radios
> without exceeding 15.4W poe.
>
> So, I take it all back. I'm good with the 1142s. Our configs should drop
> right in, or with minimal changes (interface names? Probably the same).
>

I have some AIR-AP1142N-A-K9 at the office in Santa Clara which I
provisioned with the Arista 1252.ini that I derived from the one we use at
the IETF.  Other than the weird autoprovisioning failure (losing the L3
address on BVI1), which I do not attribute to it being an 1142, it Just
Worked(tm), and all the tools (netdisco, rancid, et al) work the same too.

  Bill

>
> Chris.
>
> >> While I don't have numbers, I expect the 1252s to significantly
> outperform the 1142s in complex environments, especially with lots of
> high-speed clients. We only enable the higher-speeds on the interfaces (11M
> and up), and we certainly have a complex environment.
> >>
> >> The faster CPU/more memory/faster wireless cards/radios may give slight
> improvements as well, but I'm much less certain that it's significant.
> >>
> >> So, while I'm comfortable using the 1142s for common areas, lobbies,
> bars/restaurants, I would really like the 1252s for the meeting spaces,
> terminal room, and NOC.
> >>
> >> Plus, we're using this installation over two meetings, so the
> additional installation effort and shipping costs are worth it, in my
> opinion.
> >>
> >> Chris.
> >>
> >>
> >> --
> >> Chris Elliott
> >> CCIE # 2013
> >>
> >> “You and I are mirages that perceive themselves”
> >> --Douglas Hofstadter
> >>
> >> On Feb 7, 2013, at 11:46 AM, Warren Kumari <warren at kumari.net> wrote:
> >>
> >>> On Feb 7, 2013, at 11:28 AM, Jim Martin <jim at daedelus.com> wrote:
> >>>
> >>>> Gentlepeople,
> >>>>    A question from Rick. I personally am a bit leery, but I defer to
> our wireless experts....
> >>>>
> >>>>    - Jim
> >>>>
> >>>> Begin forwarded message:
> >>>>
> >>>>> From: Rick Alfvin <ralfvin at verilan.com>
> >>>>> Subject: WAPs for Orlando
> >>>>> Date: February 7, 2013 8:14:48 AM PST
> >>>>> To: Jim Martin <jim at daedelus.com>
> >>>>>
> >>>>> Hi Jim,
> >>>>> What would you think about using our 1142s for Orlando instead of
> the 1252s? IETF would save a lot of money on shipping
> >>>
> >>> Hmmm… I think that if we do the 1142s[0] we should *also* have the
> 1252s available, sitting in a box incase we find some critical issue...
> >>> For our new kit / future meetings we could save the shipping, etc but
> for the first (IETF) run I think having the backup parachute might be good…
> (of course, not my $$$, etc)
> >>>
> >>> W
> >>> [0]:  Which sounds like a fine idea to me, they seemed to work well in
> NANOG, better POE, etc, but not my call as I'm not the chappie supporting
> them…  Of course, one of Fabiha's jobs is attaching the external antennas
> to the 1252s, so if we don't ship them, she has less reason to attend….
> >>>
> >>>
> >>>
> >>>
> >>>>> and eliminate the hassle of dealing with external antennas. The
> 1142's have proven to be very robust for us and you'd have another
> performance data point when deciding what to replace the 1252's with when
> the gear is refreshed.
> >>>>> Thanks for your consideration.
> >>>>> -Rick
> >>>>>
> >>>>> Rick Alfvin
> >>>>> VP Business Development
> >>>>>
> >>>>> Verilan, Inc.
> >>>>> 7327 SW Barnes Rd. #215
> >>>>> Portland, OR  97225
> >>>>>
> >>>>> Mobile: +1 (585) 781-0952
> >>>>> Email: ralfvin at verilan.com
> >>>>> Skype: ralfvin.verilan
> >>>>> Website: www.verilan.com
> >>>> _______________________________________________
> >>>> ietf86-tech mailing list
> >>>> ietf86-tech at daedelus.com
> >>>> http://www.daedelus.com/mailman/listinfo/ietf86-tech
> >>> --
> >>> Do not meddle in the affairs of dragons, for you are crunchy and taste
> good with ketchup.
> >>>
> >>>
> >>>
> >>> _______________________________________________
> >>> ietf86-tech mailing list
> >>> ietf86-tech at daedelus.com
> >>> http://www.daedelus.com/mailman/listinfo/ietf86-tech
> >> _______________________________________________
> >> ietf86-tech mailing list
> >> ietf86-tech at daedelus.com
> >> http://www.daedelus.com/mailman/listinfo/ietf86-tech
> >
> _______________________________________________
> ietf86-tech mailing list
> ietf86-tech at daedelus.com
> http://www.daedelus.com/mailman/listinfo/ietf86-tech
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.daedelus.com/pipermail/ietf86-tech/attachments/20130208/84d42b4b/attachment.html 