Industry Forums → Wireless Service Providers > IT consultant pissing match?

Re: IT consultant pissing match?

Consultant is really just an industry term for someone who knows what they learned in a CCNA Boot Camp but they have no real world experience.

I usually walk away from them because they end up convincing our mutual customer that their hair brained idea will work. When it comes time to implement the hair brained idea they want me to make it work. Since I refuse to do things half ass or make things more complicated than they need to be I just decline to do it.
--
I do not, have not, and will not work for AT&T/Comcast/Verizon/Charter or similar sized company.

I've dealt with hundreds of consultants and 99.9% cant do what they recommend. They usually obsess over stupid shit while overlooking thing that actually matter. They almost always have huge egos and are very quick to tell you how well they are qualified they are. In my experiences if they have to tell you how smart they think they are the are usually not.

As a consultant you only have to be smarter than the guy that hired you because the guy that hired you is clueless about what he hired you for. He has no real way to know if you are full of shit or if you know what your talking about.
--
I do not, have not, and will not work for AT&T/Comcast/Verizon/Charter or similar sized company.

Well,

We are meeting tomorrow to start work.

I am going to be decommissioning my wireless 100/5 link, and configure the new 100/5 connection in the building. (No difference imo)

While the "other guy" here is going to install 20 N300 wireless routers to each room and configure his "RIP".

The problem I see is adding more wireless access points will encourage more users to use their wireless devices on the network instead of using 4g/3g. Seeing great WiFi signal will be deceiving in that the backend is only going to get congested with all the connections.

So 80 students, roughly 3 devices each, phones, laptops, game consoles, TVs, iPods, all on 1 100mb/5mb docsis cable modem connection.
240 devices, if each device has 10 open connections at one time, that is 2400 connections. While they will not all be online at once, some devices may do upwards of 100 or more.

I told the owner that I feel it will only get worse and advised him we need a symmetrical 50/50 via fiber or similar, but time will only tell!

It sounds to me like this setup needs to be transitioned over APs that connect to a central AP controller in stead of 20 standard routers. If the RF signals are going to overlap, it won't matter if they are using RIP. The RF channels being congested will slow things down more than routing. Since managed AP's are RF aware of each other and the client's connected to them, they can better manage the load. The setup may be able to get by with less APs than 20 and do a better job. Balancing the load, handing off to another AP as needed, ect.

I know this may be more expensive but no one has brought it up yet. I'm just looking at how this in done in dense networks like university campuses. Here they plan for about 30-40 clients per AP but new firmware and controllers will support more. Here they are using Aruba Networks 135 AP to replace the Cisco ones. These run about a thousand a pop though for the AP. I'm sure the controller is not cheap either but you don't need one to support thousands of APs like a University.

»www.arubanetworks.com/products/a···-series/

An unmanaged RF space will be the kiss of death...especially at this density and proximity.

Ruckus performs well in this environment...pricewise in the ballpark of the Aruba gear.

I'd argue that the extra cost for the controller is worth it, since it manages the system automatically, and (in theory, and actually) just works. I'd say if you had 12+ APs in your system, it'd be worth the cost for the controller, for a variety of reasons.
--
Many believe, but few will question...I decline to be blind.

Well, after you've spent $12K for a high-performance managed AP system, it hardly makes sense to not spend the additional 10% to have the controller tie it all together. I'm a tinkerer as much as the next guy, but in a production environment, not so much. Besides, while you're fiddling with settings and such, other things are not getting done.

At any rate...I would, in the words of Ron Popiel, just as soon set it and forget it.

On the other end of the spectrum, so to speak, how about using many micro APs (in the OPs case with 20 additional) that only service the immediate room? Who makes the smallest, lowest powered full function AP? How low can one go? Is there a USB dongle AP available? Just thinking out loud here...
--
Many believe, but few will question...I decline to be blind.

Another option:

»www.handlink.com/products_wap-001.php

I build lots of wireless networks in MDUs and use Ruckus (with Zone Directors) almost exclusively. That being said, I have found we get better performance and lower subscriber churn by wiring each unit and then installing a dedicated wireless router for each customer. We use UBNT AirRouters with a custom self-provisioning firmware which does 802.1x auth.

Ruckus has one as well:
»www.ruckuswireless.com/products/···oor/7025

So, I dunno, Steve... Sure your job is to help the customer, but is it to help your competition? I think not. Maybe if the customer wants to pay me to help the competition, then sure but otherwise maybe not so much. Then take into consideration that the man your customer wants to replace you with spells throttling as troddling and talks about side loops... I dunno... What do I know? I never wrote a book. Well, I never wrote a book you will be allowed to read.
--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."