|
Home > Archive > CWNP > July 2003 > Baseline statistics needed!
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Baseline statistics needed!
|
|
| meijin 2003-07-23, 11:00 pm |
| Hey all...I have posted this once before and I don't believe anyone took a swipe at it at all. I am still in need of this and hope someone will jump in here.
I need to come up with a set of baseline statistics to use in comparison against real life statistics on the installation of a new WLAN.
For instance, when testing it, what percent of packets and be dropped? What percent can be lost? Collision percentage? BER? What other statistics?
I need to develop a baseline and can find nothing in any of my reference books or on the net to work with. I am in a bit of a bind here and would appreciate some help here.
Thanks! | |
| wirelessboy 2003-07-24, 1:21 pm |
| this is a tough one meijin
today between two Aironet-350 i got 5% signal, but i was able to do a video conference!
what to beleive?
i stopped beleiving in these % values, the moment the 802.11B modems came into the market with their attractive GUI based software, i would rather prefer a HP-Invent Meter.
anyway, i will collect all my data and arrive at a figure soon.
regards | |
| meijin 2003-07-24, 2:04 pm |
| You would think that there is some sort of network or industry standard...or at least I think so.
I just need to come up with something to use as a baseline as far as comissioning a new wireless network goes.
Thanks and I look forward to hearing from you.
FYI - I am about to start doing alot of travel over the next 3 weeks...so if you don't hear much from me don't think I am ignoring you! | |
| Devinator 2003-07-26, 10:11 pm |
| That's one of the fun things about WLANs is that your baseline is different for every installation! :0) Every facility is different, every environment is different, each mfr's gear is different, etc. Because of these reasons, there is no such thing as a "standard baseline." Your performance baselines will be determined once you've installed the WLAN, not before. For this reason, your site survey has to be good. Mfr's are shy about telling you "Factor A should be less than X% and Factor B MUST be more than 45%", etc because everything in WLAN networking is based on the environment. Knowing the relationships between retransmissions, signal strength, throughput, and other factors gives you the mental tools to make educated guesses - which is as good as it gets most of the time. ;-) | |
| meijin 2003-07-27, 12:43 am |
| Devin:
I can see your point and can agree with it, for the most part...
We will be using the same hardware all of the time with just the exception of possibility of doing a "one off" with Cisco APs.
So, based on your experience, what is the criteria that you would use for comissioning a WLAN? The customer needs to be able to see some sort of performance level. That is what I am trying to get to.
Am I making any sense at all?
Thanks! | |
| Devinator 2003-07-27, 4:13 pm |
| What you're describing is an application analysis. This is a procedure where you go to the customer's site, interview the network manager, gather all pertinent data, and then test the environment to verify that a wireless LAN can meet the customer's needs and expectations. There's no pre-set baseline for this - each install is different. If you're on the mailing list, you can have access to our PDF downloads. One of them is the Site Survey Network Manager Interview form. Part of the final site survey form that you will give to the customer will be an application analysis form that shows that the wireless LAN, once installed, will be able to handle their traffic, mobility, and security needs. ;-) | |
| meijin 2003-07-27, 4:48 pm |
| I see what you are saying and this is part of the reason why I like NetIQ Chariot as you can, essentially, build a traffic pattern and see the results (in use with a sniffer as well if you want).
So, again, (or still) I am curious as to the process. If you have say...20 users in a sector, it would be a bit difficult to try and get all 20 up and running and testing at the same time. So, do you use something like Chariot or do you use something else?
Thanks! | |
|
|
|
|
|