>But for the >price of Ubiquiti gear it seems very interesting to investigate what >could be done.
Well, thats the golden question... We dont currently use Ubiquiti yet in a live network, but we cant ignore the value proposition. When APs are $90, do we need APs that scale? But we do need radios that stay associated though. Good to hear, some are reporting the new beta5 firmware is appearing to run stable with WDS. Tom DeReggi RapidDSL & Wireless, Inc IntAirNet- Fixed Wireless Broadband ----- Original Message ----- From: "Rubens Kuhl" <rube...@gmail.com> To: "WISPA General List" <wireless@wispa.org> Sent: Tuesday, April 13, 2010 11:33 PM Subject: Re: [WISPA] Ubiquiti AirOS Comparison On Tue, Apr 13, 2010 at 9:00 PM, can...@believewireless.net <p...@believewireless.net> wrote: > After some large "experiments" with Ubiquiti...... Canopy 430 here we > come! Too many problems with latency and WDS re-reg issues. These > seem to work pretty well for PtP links, but PtMP is just terrible for > VoIP or anything that can't handle latency spikes. Is it a coincidence that your e-mail address is canopy@ ? :-) I've noticed WDS problems with Wi-Fi gear for a long time; previously it was more easily triggered by security protocols and channel selections. With 802.11n there seems to be added problems with association and performance impacts on the 802.11n MAC aggregation mechanisms. That's why I currently believe in scaling Wi-Fi based (even with proprietary polling protocols like nstreme or AirMax) without using WDS, which is not part of the 802.11 specification BTW. One can add a Mikrotik RB-750 at every customer site for US$40 and achieve whatever Layer-2 transparency (by using MPLS/VPLS, EoIP, Ethernet over PPP) and user enforcement/control (filtering to allow only PPPoE frames, doing a hotspot authentication at the RB750 or what fits best your business model) and then use whatever radio network is offering good quality at good prices at that time. Regarding the latency spikes, 802.11e might be useful and probably more powerful having a CPE device that could mark QoS/ToS/DSCP/CoS/EXP before it comes to the radio. I haven't seen a working 802.11e-based network yet, but there are very few end-to-end QoS-enabled IP networks on the world and it took $M, not $k, money to build them, so it was very unlikely that I could find one with 802.11 devices. But for the price of Ubiquiti gear it seems very interesting to investigate what could be done. Rubens -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/ -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/
