95th Percentile != Lame

Mon Jun 4 03:41:03 UTC 2001

On Sun, 3 Jun 2001, Richard A. Steenbergen wrote:

> On Sun, Jun 03, 2001 at 07:42:13PM -0700, David Klindt wrote:
> >
> > > I still fail to see how "peak bits" or "bursted bits" are more
> > > expensive than "regular bits".  A 100Mbit FE port costs whatever it
> > > costs, and does not fluctuate with usage.  This is true of almost all
> > > of your links within the network - excluding those where you have
> > > negotiated usage-based billing.  An OC3, point to point, costs as much
> > > as it costs irrelevant of its usage.  Therefore, every bit that
> > > crosses this circuit has a cost.
> > >
> > > Why not simply pass this cost on to the customer bit for bit?
> >
> > It is NOT that the each bit has the same cost - it is the cost of
> > maintaining enough EXTRA bandwidth so that the downstreams do not
> > bounce up against the ceiling. That amount is basically covered by
> > using the 95 rule.
> 
> But peak vs non-peak has little to do with 95th percentile. 

Sure they do. I sell bandwidth. I either place a limit on each port, or I
let a client go full open - their call. I MUST be in a position to cover
those costs and yes, at times unused bandwidth. That cost must be past on
to the client if I am to remain in business.  If all clients were willing
to set a ceiling and be forced to live within that ceiling, then no
problem. Clients who select a ceiling pay for the (100 percent) of that
bandwidth (ceiling).

If I do not have the bandwidth to cover the peaks of all clients at the
same time, I am shorting the clients.