5 GHz was only usable in hallways (deserted) and the front part of suites (shower and restroom). Where users are present we were measuring -80 dBm or less at 5 GHz. That dorm doesn't have common areas. In common areas, it would make sense to use 5 GHz of course. The extra capacity is provided by providing smaller cells at 2.4 GHz.
The other problem with weak 5GHz and stronger 2.4GHz is that devices (e.g. Macs!) do not join 5 GHz, even if available. Tricks like Bandwidth Steering can help this behavior, but not if the difference is too large between the 2 spectrums. BTW: If a large majority of users start watching Netflix in restrooms we will have to swap APs! Philippe On Mar 22, 2011, at 4:26 PM, David J Molta wrote: Wasted in what sense, Philippe? Residence halls are obviously high-density environments so capacity is a big concern, especially during peak usage periods. Even if the 5 GHz 11n channel can’t provide full coverage for the area under consideration, if it offloads even 25-30% of the 2.4 GHz 11n traffic, it seems like it would be worth the extra cost because it would result in better performance under heavy contention for both 2.4 GHz and 5 GHz users. Dave Molta On 3/22/11 3:59 PM, "Hanset, Philippe C" <phan...@utk.edu<x-msg://2463/phan...@utk.edu>> wrote: Having done a Dorm Installation last week, let me add another point: 5 Ghz is great, but in some places you might want to skip the expense. We had no choice but to feed Dorm suites from the center hallway. After a thorough testing we came to the conclusion that we would only provide 802.11n at 2.4 GHz since too much of the 5 GHz signal was wasted. So instead of using Aruba AP-105 ($695 list) everywhere in that building, we settled for Aruba AP-93 ($395 list). With the savings, we did smaller cells to somewhat compensate for the lesser capacity. Each bedroom still has an ethernet drop (unfortunately unusable for APs for architectural reasons). Philippe Univ. of TN On Mar 22, 2011, at 3:38 PM, heath.barnhart wrote: If nothing else, you will have a documentation showing what your coverage is and can uncover any gotchas. If someone says they are having issues in an area, you pull up the survey and have instant access to more information to help uncover the reason behind their issues. Heath On 3/22/2011 2:11 PM, John Kaftan wrote: So I hate to dig this up again but nobody really responded to Jeff Sessler’s post “Given the need for designs based on capacity rather than coverage, do those who've done site surveys previously feel they are still worth the trouble?” Seems to me wireless surveys are for determining coverage which is something we can easily measure. We can require that an area will have no less than -68 dBm signal and do the survey to determine what it will take. However, if folks are saying that in a high density area like a ResHall just providing coverage is not enough and we must go much denser what good is the survey? If coverage is not enough then how do we determine our density? Is it just by feel? Up until now I figured I was not going to do a survey. I figured for the cost of the survey I could buy an additional 30-50 APs. When pulling wire I’d have facilities leave a 20’ coil and pull double the wire I originally guessed based on past experience. Then we would just “Throw it up” and see what happens. If we move slowly and do a ResHall at a time we should be able to get a feel for it. Now I have a shot at doing a survey this summer after the fact by using students from a nearby University that has a MS in Networking as an internship. The cost is much less than a professional survey but I have to ask if it is still worth it if capacity is what we are going for? Perhaps I should be looking at a different internship. There is certainly plenty to do around here. John Kaftan Infrastructure Manager Utica College 315.792.3102 From: The EDUCAUSE Wireless Issues Constituent Group Listserv [mailto:WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU] On Behalf Of John Kaftan Sent: Wednesday, March 16, 2011 8:16 PM To: WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU<x-msg://2463/WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Wireless Site Survey cost I have everyone held back to 2 Mbs on wireless. That seems to be a good number for now. Nobody is complaining and it helps to keep their experience consistent. They can watch a Netflix movie with that. I imagine Netflix would use more bandwidth if it could. I have not tested though. On 3/16/2011 6:28 PM, Brian Helman wrote: If people are building new dorms, I’d definitely run copper to any common rooms if you support any gaming consoles. Honestly though, we have a good density of wiring even in the dorms and I’m pretty close to shutting down or at least limiting the bandwidth available for video on the wireless network. Netflix, Flash and Youtube are killing it (not to mention our Internet connection). -Brian From: The EDUCAUSE Wireless Issues Constituent Group Listserv [mailto:WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU] On Behalf Of Joel Coehoorn Sent: Tuesday, March 15, 2011 10:30 PM To: WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU<x-msg://2463/WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Wireless Site Survey cost Agree I wouldn't run new "port-per-pillow" drops, but I wouldn't ditch existing drops (just update the switching) and anywhere you have apartment-style living I would put a wired port in the common space for game consoles/blu-ray/smart tvs/etc. Those who actually use the ports will be the few who know enough to know why it's better, and they also tend to be your heaviest users. It's nice to get some of the gaming and netflix traffic out of your airspace. On Mar 15, 2011 7:50pm, John Kaftan <jkaf...@utica.edu<x-msg://2463/jkaf...@utica.edu>> <mailto:jkaf...@utica.edu> wrote: > Thanks, but I have purchased already. We will be doing this backwards. We > are pulling extra drops and leaving 20' coils of cable above the ceilings and > then throw up the APs and see what happens. Not perfect but we have been > doing alright with that. We have a feel for it and the students report > happiness. This summer we will do the survey to tighten things up a bit. I > am considering dropping the wired ports as our LAN is past due for a refresh > and I do not want to re-invest in the port-per-pillow model. > > > > John > > > > On 3/15/2011 7:09 PM, Brian Helman wrote: > > > Have you already selected a wireless product? If not, I think you'd be far > better served issuing an RFP for full procurement and installation, with > signal guarantees (I'd recommend -68dBm). If you have holes, the contract > should be on the hook for it. Take advantage of this economy. Vendors will > jump on this. > > > > Remember, antennas vary GREATLY. If you do a survey and then bid out and end > up with a different product than you conducted the survey with, you could end > up with holes. > > > > -Brian > > ________________________________________ > > From: The EDUCAUSE Wireless Issues Constituent Group Listserv > [WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU<x-msg://2463/WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU>] > On Behalf Of heath.barnhart > [heath.barnh...@washburn.edu<x-msg://2463/heath.barnh...@washburn.edu>] > > Sent: Tuesday, March 15, 2011 9:57 AM > > To: > WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU<x-msg://2463/WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> > > Subject: Re: [WIRELESS-LAN] Wireless Site Survey cost > > > > If you have any resellers/technology partners/consultants you might ask them. > Standard consultant fees would probably apply (I've seen $150-$300/hour). If > they're good they should be able to survey a couple buildings in a day (which > should be less than $1500 a floor). You could also do it yourself. Someone > mentioned Ekahau; we use Airmagnet Survey. Its good too have a survey > solution for troubleshooting anyways. > > > > > > -- > > Heath Barnhart, CCNA > > Network Administrator > > Information Systems and Services > > Washburn University > > Topeka, KS 66621 > > > > > > On 3/14/2011 4:46 PM, Winston Chow wrote: > > Usually companies don't like to do site surveys because they do it assuming > you'll buy APs from them. If anything I found that companies will do it for > a lot of money but give you a significant credit if you buy > APs/controllers/service from them. > > > > That doesn't work with our procurement system that needs 3 lowest bidders. > > > > Good Luck! > > > > -Winston > > > > > > > > On Mon, Mar 14, 2011 at 6:41 AM, John > Kaftanjkaftan@utica.edumailto:jkaftan<x-msg://2463/Kaftanjkaftan@utica.edumailto:jkaftan>@utica.edu>> > wrote: > > I know this is a crazy question with tons of variables but I am trying to at > least get an idea of what it would cost to do a wireless survey in our > residence halls. We have 7 buildings built over the years with a variety of > construction materials. Each building has 3-4 floors. We have a total of > 1100 students living on campus. > > > > Has anyone had a commercial wireless survey done and if so can you give me > any idea of what I would be looking at? > > > > My intention is to do this via an Internship so I do not really want to shop > this out and put vendors through the paces. I just want to give an estimate > of what it would cost the college if we were to have a commercial provider do > the work. > > > > John Kaftan > > Infrastructure Manager > > Utica College > > 315.792.3102 > > ********** Participation and subscription information for this EDUCAUSE > Constituent Group discussion list can be found at > http://www.educause.edu/groups/. > > > > ********** Participation and subscription information for this EDUCAUSE > Constituent Group discussion list can be found at > http://www.educause.edu/groups/. > > > > > > > > -- > > Heath Barnhart, CCNA > > Network Administrator > > Information Systems and Services > > Washburn University > > Topeka, KS 66621 > > > > ********** Participation and subscription information for this EDUCAUSE > Constituent Group discussion list can be found at > http://www.educause.edu/groups/. > > > > ********** > > Participation and subscription information for this EDUCAUSE Constituent > Group discussion list can be found at http://www.educause.edu/groups/. > > > > > ********** > > Participation and subscription information for this EDUCAUSE Constituent > Group discussion list can be found at http://www.educause.edu/groups/. > > ********** Participation and subscription information for this EDUCAUSE > Constituent Group discussion list can be found at > http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.
