wireless network

[kevin_lane]

what do people use for their wireless infrastructure , how well does it perform, what would you change about it

[m25man]

Ubiquiti Unifi,
Brilliant,
Nothing....

Deploying UniFi - Ubiquiti Networks Forum

The big guys are running scared and its long overdue.

[Dos_Box]

Moving to Networks forum.

[john]

Ruckus from @Net-Ctrl
Performs great, had it in all the schools I have worked in and its been no problems and the same goes for the support from the team at Net-Ctrl

What would I change about it.... Bring back Douglas the dog (sorry Edugeek joke and long story!) no seriously, I don't think there is anything at this moment I would change with it as it just works and does as it says, and Ruckus are very happy to take feedback on the product and do so as I have seen over the last 4 - 5 years its been in the UK

[MatthewL]

Aruba - its great works well, only thing I would change is to revamp our whole network and start from scratch as its shocking. One thing I would make more use of is their RAPs if we could afford more of them.

[EduTech]

I also recommend the Ruckus equipment, I recommend you speak to @Net-Ctrl for further information as they have a wealth of knowledge on wireless as a whole and will be able to provide you with the best solution for your environment.

I have deployed Ruckus Wireless to Schools & Businesses and it has not failed me yet from 5 APs Including a External Site to Site Bridge to 25+ APs

@john - Yes Douglas should come back he seems to have gone a bit quiet! ;-)

James.

[mavhc]

Cisco
Well
Enabling 802.11n wasn't quite as simple as it should have been

[kevin_lane]

brill think Ruckus might be the ticket then "hopefully" thanks people

[netadmin]

We use Ruckus. It really works great, although we purchased the single band "N" access points; if I had to do the install again, I would have gone with the dual band "N" access points to give a bit more speed and more capacity for the future.

[psydii]

Dual Band, Dual Radio APs that support at least 3x3 MIMO
Ideally 3x3 (or better) Wireless clients.
A metal roof to create viritual APs in the sky (just lucky there)
A support contract with the manufacturer.
Certified structured cabling between your APs and the rest of the network
A site survey to check there's no nasty RF interference on site.
End users devices purchases specification requirement that they must work with {insert wireless vendor here}
Handover documentation and training.

What would I change? More radios from day one. We planned for blanket coverage with some areas having better signal than others, to be augmented over a number of years with additional AP's based on device usage around the site. In practice we should have bitten the bullet and gone at least one radio per classroom (in our case that would have physically been one AP per two classrooms).

I've never yet had a controller fail in the seven years I've been running controller based WLANs, but having them as a single point of failure for what is now in effect THE network infrastructure worries me. Even with a 4hr swap out contract.

[m25man]

Sorry but...

With Broadcom manufacturing .ac chips, handset vendors already planning integration and AP manufacturers already lining up kit, unless your controllers and AP's are upgradable to as yet un-established standards anything on the current vendors shelves is likely to be obsolete before you have even managed to get part of your ROI....

A few key design enhancements in 802.11ac compared to 802.11n are
1. Wider channel width i.e. 80 MHz mandatory, or 160 MHz optional (40 MHz maximum in IEEE 802.11n)
2. Higher modulation i.e., 256 QAM (64 QAM in case of IEEE 802.11n)
3. MU-MIMO (Multi User MIMO, SU-MIMO in case of 802.11n).
4. Short GI which is 400 μsec
5. STBC (Space Time Block Coding) and LDPC (Low Density Parity Check) codes
6. Maximum up to eight spatial streams (four in IEEE 802.11n)

The SAI 802.11ac PHY includes support for these features:
- Support for all channel combinations including 160 MHz contiguous as well as 80+80 MHz combination
- Full PLCP header support for VHT mode including VHT-SIG-A/B as well as VHT-LTF and VHT-STF
- PHY configuration and legacy protection in mixed mode – including CTS/RTS and L-SIG-TXOP
- All MCS and FEC modes defined by the standard
- Multiple antenna support in a number of modes including

o NxM MIMO combinations (Up to 8x8) with no restrictions on N and M
o Spatial mapping and STBC support
o All CSD combinations
o SU and MU MIMO

- Receiver supports includes channel estimation and MIMO detection
- Beamforming support includes

o Steering vector calculations using antenna mesaurements
o Implicit/Explicit beam forming reports
o VHT compressed beamforming report
o MU-MIMO beamforming report
o MIMO beamforming algorithms using received reports

- Performance data for this is measured

o Using channel definitions from the 802.11ac standard (in draft form) models a-f
o Delivering performance data such a BER/PER as well as RSSI and Channel frequency response per link.

The next 18 months in the WiFi market is going to get really interesting, that's for sure. I for one will be sure not to go spending £000's on obsolete kit.

[m25man]

A very interesting video from the wifi mobility symposium today, the Ruckus guy definately not comfortable with the comments of his peers.
If you have 47mins to spare its worth a view.
Gigabit Wi-Fi (802.11ac/ad) Discussed at the Wi-Fi Mobility Symposium on Vimeo

[steveg]

We've got Ruckus kit in 2 schools and love it - it just works. Have deployed netgear git (managed) in 2 other schools, and it works ok, and unmanaged netgear kit in countless others, which also works fine for smaller deployments.

We've just purchased 6 Unifi access points, and are about to do a strait swap in one of the managed netgear schools as a trial. I will do some tests before and after to see how it compares. - The price on this kit seems too good to be true - 6 ap's and a controller (ok software only so you have to run it on a vm or old pc) for £400!

Steve

[Soulfish]

Big fan of Aerohive here. Just works, and no controller to worry about

[gshaw]

Keep the replies coming guys, I'm considering Ruckus, AeroHive and now Ubiquity (due to the crazy low price)...

Gut instincts say...

• Ruckus - proven and nice solution, good performance and seems to cover with less APs than some competitors... just the cost for controller that's a bit of a pain
• AeroHive - nice management, like the controller-less solution but don't like the fact the costs are just shunted onto "management" which becomes an ongoing cost (or an upfront VM purchase that's the same price as the Ruckus controller!)
• Ubiquity - so cheap I could cover the whole building for the same price as 1 controller for other solutions but I have some doubts over long-term reliability and build quality (it's almost too good to be true so have corners been cut somewhere?)