[148504] in North American Network Operators' Group
Re: enterprise 802.11
daemon@ATHENA.MIT.EDU (Chris Burwell)
Mon Jan 16 21:44:02 2012
In-Reply-To: <-7665060707062421807@unknownmsgid>
From: Chris Burwell <cburwell@gmail.com>
Date: Mon, 16 Jan 2012 12:44:22 -0500
To: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
I used Xirrus before about 2-3 years ago. They are great for
addressing density issues without adding a large amount of APs in one
area. As with any wireless solution, it does have it's limitations. In
our case the building was very challenging with solid concrete walls
on top of lockers on each wall. The number of APs (IIRC they call them
arrays) needed didn't really save us much in the end. One thing I did
not like is that you had to use their power injectors because, at the
time, their arrays needed more power than any switch could provide.
Pricing ended up being the ultimate fall of Xirrus in our environment.
To get the coverage that we needed (real world) they were considerably
more than the early HP/Colubris solution that we ended up with.
- Chris
On Sun, Jan 15, 2012 at 6:42 PM, Mike Lyon <mike.lyon@gmail.com> wrote:
> Another one which looks promising for high-density locations is Xirrus
> (www.xirrus.com)
>
> Haven't ever used them though.
>
> -mike
>
> Sent from my iPhone
>
> On Jan 15, 2012, at 15:36, Greg Ihnen <os10rules@gmail.com> wrote:
>
>> Since we're already top-posting=85
>>
>> I've heard a lot of talk on the WISPA (wireless ISP) forum that 802.11g/=
n starts to fall apart with more than 30 clients associated if they're all =
reasonably active. I believe this is a limitation of 802.11g/n's media acce=
ss control (MAC) mechanism, regardless of who's brand is on the box. This i=
s most important if you're doing VoIP or anything else where latency and ji=
tter is an issue.
>>
>> To get around that limitation, folks are using proprietary protocols wit=
h "polling" media access control. Ubiquiti calls theirs AirMax. Cisco uses =
something different in the "Canopy" line. But of course then you've gone to=
something proprietary and only their gear can connect. So it's meant more =
for back-hauls and distribution networks, not for end users unless they use=
a proprietary CPE.
>>
>> Since you need consumer gear to be able to connect, you need to stick wi=
th 802.11g/n. You should limit to 30 clients per AP. You should stagger you=
r 2.4GHZ APs on channels 1, 6 and 11, and turn the TX power down and have t=
hem spaced close enough that no more than 30 will end up connecting to a si=
ngle AP. 5.8GHz APs would be better, and you'll want to stagger their chann=
els too and turn the TX power down so each one has a small footprint to onl=
y serve those clients that are nearby.
>>
>> Stay away from "mesh" solutions and WDS where one AP repeats another, th=
at kills throughput because it hogs airtime. You'll want to feed all the AP=
s with Ethernet.
>>
>> Greg
>>
>> On Jan 15, 2012, at 4:22 PM, Nathan Eisenberg wrote:
>>
>>> Ubiquiti's Unifi products are decent, and have *MUCH* improved since th=
eir original release (amazing what you can do with better code!). =A0In the=
original release, you had to have a management server running on the same =
L2 network as the Aps - they've moved the management to a L3 model so you c=
an put the controller elsewhere. =A0The big PITA with their system is that =
any change requires 'reprovisioning' the APs, which means rebooting all of =
them in sequence. =A0They've added VLANs, multiple SSID's/AP, wireless back=
haul/chaining, guest portalling, and limiters to balance the # of clients /=
AP.
>>>
>>> In a noisy environment, I've found that they top out at around 30 devic=
es / AP for good performance, and 50 devices / AP for 'working/not working'=
. =A0In a clean environment, I've seen decent performance with 70 - 100 dev=
ices / AP. =A0Of course, if one bad client comes along (with a card that do=
esn't backoff its TX power, etc), it can wreak havoc with higher densities.=
=A0You really can't argue with Unifi's price.
>>>
>>> If you move up the price scale, Meraki seems to be a good midrange solu=
tion, and they have some really sweet reporting functionality. =A0They're m=
ore expensive, though.
>>>
>>> And then, yes, Cisco is the gold standard, but it will cost you some go=
ld to get it.
>>>
>>> Nathan
>>>
>>>> -----Original Message-----
>>>> From: Mike Lyon [mailto:mike.lyon@gmail.com]
>>>> Sent: Sunday, January 15, 2012 11:54 AM
>>>> To: Meftah Tayeb
>>>> Cc: nanog@nanog.org
>>>> Subject: Re: enterprise 802.11
>>>>
>>>> Ubiquity (www.ubnt.com) has their Unifi line of products. It's still p=
retty new
>>>> in the marketspace and this, working out the bugs. I use their other p=
roducts
>>>> exclusively for outdoor wireless.
>>>>
>>>> However, in the offices ive done, ive used Cisco's WLC 4402 controller=
which
>>>> supports 12 access points. They have controllers which support more AP=
s as
>>>> well.
>>>>
>>>> Hit me up offlist if you have any quesrions.
>>>>
>>>> -mike
>>>>
>>>> Sent from my iPhone
>>>>
>>>> On Jan 15, 2012, at 11:39, Meftah Tayeb <tayeb.meftah@gmail.com> wrote=
:
>>>>
>>>>> Ubiquity
>>>>> or ubikity, maybe is miss spelled
>>>>> Someone correct the spelling for him please thank you
>>>>> ----- Original Message ----- From: "Ken King" <kking@yammer-inc.com>
>>>>> To: <nanog@nanog.org>
>>>>> Sent: Sunday, January 15, 2012 9:30 PM
>>>>> Subject: enterprise 802.11
>>>>>
>>>>>
>>>>> I need to choose a wireless solution for a new office.
>>>>>
>>>>> up to 600 devices will connect. =A0most devices are mac books and mob=
ile
>>>> phones.
>>>>>
>>>>> we can see hundreds of access points in close proximity to our new of=
fice
>>>> space.
>>>>>
>>>>> what are the thoughts these days on the best enterprise solution/vend=
or?
>>>>>
>>>>> Thanks for your replies.
>>>>>
>>>>>
>>>>> Ken King
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> __________ Information from ESET NOD32 Antivirus, version of virus
>>>> signature database 6793 (20120113) __________
>>>>>
>>>>> The message was checked by ESET NOD32 Antivirus.
>>>>>
>>>>> http://www.eset.com
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> __________ Information from ESET NOD32 Antivirus, version of virus
>>>> signature database 6793 (20120113) __________
>>>>>
>>>>> The message was checked by ESET NOD32 Antivirus.
>>>>>
>>>>> http://www.eset.com
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>
>
