If you're going to be really measuring yourself against an SLA, the only sane
way I can see would be to do so with a metrics tool like 7Signals, Streetwise,
or NetBeez. This would also force you to specify exactly *where* you're
trying to offer a given SLA, allowing you to focus on critical areas like
classrooms, while explicitly leaving unmonitored areas like parking lots at a
best-effort only level.
Once you have metrics being gathered via an objective tool, questions like one
AP out of a group going offline don't matter nearly so much anymore. As long
as you're hitting your metrics (able to connect, at least XX throughput, no
more than YY jitter) you're good, regardless of individual component level status.
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 8/26/2015 2:14 PM, Julian Y Koh wrote:
On Wed Aug 26 2015 13:02:49 CDT, Hunter Fuller <hf0...@uah.edu> wrote:
Of course I can't speak for everyone, and I don't know that I would
lay out an SLA saying wireless will be up 99.999% of the time or
anything, but it just doesn't seem as fragile as one might think
initially.
The next problem is uniformly defining uptime for an enterprise/campus-wide
service. How does the failure of a single AP (or a single switch if you're
talking wired networking) impact your uptime number? The issue is even more
nebulous for wireless since I would bet in most cases if you have a single AP
out in most of our buildings, there's still some residual connectivity
available from other nearby APs.
**********
Participation and subscription information for this EDUCAUSE Constituent Group
discussion list can be found at http://www.educause.edu/groups/.
