We're about six months into our initial foray into 7signals, and I think that
Norman pretty much hit it head on. He's especially correct in the value in
the initial professional services, in terms of learning the tool and
understanding the data. It has the ability to show you a wide range of views
of a lot of different data sets, and some of the ones they focus on in their
analysis aren't necessarily ones that you would have thought to track on your own.
Even more importantly, though, I don't think you can overemphasize what a big
change it is to have detailed, objective feedback on the impact of
configuration changes. Their analyzer UI allows you to stick change events
into the database, which will then show up as markers the reports, allowing
you to easily view before vs after affects. Getting this kind of data after
tweaking wireless knobs and dials is a world of difference vs trying to read
the tea leaves through ticket metrics.
Frank Sweetser fs at wpi.edu | For every problem, there is a solution that
Manager of Network Operations | is simple, elegant, and wrong.
Worcester Polytechnic Institute | - HL Mencken
On 7/10/2014 1:55 PM, Norman Elton wrote:
We were introduced to 7Signal from their presentation at the Wireless
Field Day, and seeing them on a few listserv conversations. We
currently have ~15 sensors (eyes) scattered around our network, mostly
in residence halls. There is an upfront cost, and yes, their licensing
seems a little more complicated than it really needs to be. That
aside, it is a very powerful tool. To be honest, we don't use it as
much as we should. More on that in a moment.
7Signal Eyes gather tons of statistics, some passive (signal strength,
percentage of retries in the air, etc), others collected by connecting
to the AP (TCP throughput, DNS availability). The data is passed back
to a database, where a web-based analyzer allows you to slice and dice
into graphs and SLAs.
We purchased the product for two main purposes. First, to help
investigate student complaints. If someone complains that wireless is
sluggish every night between 2:00 and 3:00 AM, we can deploy an Eye to
gather statistics. Is it just a matter of load? Airtime exhaustion?
Interference? Second, to help guide us in making tweaks to our AP
configuration. What happens when we enable 40-MHz wide channels? What
gains were made when we enabled DFS channels? How about tweaking the
beacon interval? These are all changes that theoretically help the
network, but we don't have a way to actually measure their effect.
They sell the product with a hefty professional services overhead. I
see two reasons behind this...
First, they want to show you how to use the product, then make
suggestions on how to improve your wireless network, and show you how
to verify that the improvements resulted in performance gains. None of
their suggestions were groundbreaking, but their engineers (at least
those I worked with) knew their stuff. We're reasonably intelligent
wireless folks, but they deal with tweaking networks every day. We
felt it was money well spent.
The second reason points to our major complaint with 7Signal. Their
web-based analyzer has a very steep learning curve. It's akin to
learning the US tax system by reading the IRS tax code. Sure it's
possible, but you really need a CPA to help guide you along. You're
dealing with a very powerful GUI with lots and lots of knobs, not all
of which are clearly explained or even relevant to what you're doing.
I get the impression that the product was developed over many years by
very smart wireless people who used it every day, but never bothered
to step back and look at it as a product a customer would buy / learn
/ use.
We've expressed this concern, and have been assured that they are
working on a new analyzer platform. But until it's running on our
servers, it's vaporware :-/.
So, all of that said, it's a powerful tool. We've gotten useful
information from it, and they have guided us through a number of
tweaks. We have not seen a gigantic performance improvement, but feel
our ability to monitor performance justifies the product.
I'm happy to answer other questions or jump on a call.
Norman Elton
p.s. I'm really convinced that 85% of their statistics could be
gathered by the AP directly. I'm pushing our vendor, Aerohive, to
implement SFlow with 802.11-level information. Instead of collecting
information from a few sensors, every AP could report a
statistically-meaningful amount of sampled packets. If anyone else
feels this would be useful, let's get that conversation going!
On Thu, Jul 10, 2014 at 12:27 PM, Lee H Badman <[email protected]> wrote:
Thanks, Norman. I'm thinking a lot of people are curious. I know one glaring
point is that there seems to be heavy push for professional services, and I'm
not real keen on needing multiple licenses to bring an Eye to life (seemingly
to spread costs out) but those are my personal biases. Anything you can share
would be appreciated!
-Lee
-----Original Message-----
From: Norman Elton [mailto:[email protected]]
Sent: Thursday, July 10, 2014 10:18 AM
To: The EDUCAUSE Wireless Issues Constituent Group Listserv
Cc: Lee H Badman
Subject: Re: [WIRELESS-LAN] Epitiro StreetWise, 7signal
Lee -
Just saw your email. We have 7Signal at William & Mary. Would love to
chat about our experience. Let me know if you're still interested.
Thanks,
Norman Elton
College of William & Mary
On Mon, Apr 28, 2014 at 3:14 PM, Lee H Badman <[email protected]> wrote:
Greetings to the list-
Wondering if anyone is using or has taken a serious look at (to the point
where you’ve gotten quotes) either 7signal or Eptiro StreetWise for wireless
performance measurement. Epitiro is newer to Wi-Fi game, less PR to date,
but established in mobile carrier world.
We can chat on list or off, as you’re comfortable with.
For those using either:
Which one?
For how long?
How “bad” did your network need this sort of help before you started?
Biggest win?
Any frustration?
Perspective on company, support, etc?
Now that you’re in it with them, has the investment been worth it?
Has the time commitment to get value been acceptable?
For those shopping:
Which one?
Actively considering?
Did sticker shock chase you away?
Any other thoughts or comments on the general topic welcome as well, as we
research whether to go this path.
Thanks-
Lee Badman
Syracuse University
********** 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/.
