[rrd-users] Re: disabling interpolation?

[Alex van den Bogaerdt]

Tobias Oetiker wrote:

> Also I would NOT quantize because people get confused by the data
> ... you should not fix the statistics to apeal to the masses.
> Rather take the oportunity to educate them by having a link to a
> page explaining why there were 4.7 modems in use over the last hour
> ... because at least when you get to longer intervals like yearly
> average the rounding will become sort of a liability ... 

If you want to know the average number of modems in use: yes, you're
right.  This number is very valuable for an ISP as they do not want
to waste money.

Most users do not care if the average is 4.7 or 12.1, the only thing
they care about is the maximum.  If the maximum number of modems is
reached it is *very* probable that this was the cause of that busy tone.

It is simply impossible to have 4.7 modems in use therefore it *is* a
bug if you present this number as a maximum.

Both average and maximum are equally important in this example and both
need to be as accurate as possible.  The interpolation done by RRDtool
is okay for the average, it is not for the maximum.

regards,
-- 
   __________________________________________________________________
 / alex at slot.hollandcasino.nl                  alex at ergens.op.het.net \
| work                                                         private |
| My employer is capable of speaking therefore I speak only for myself |
+----------------------------------------------------------------------+
| http://faq.mrtg.org/                                                 |
| http://rrdtool.eu.org  --> tutorial                                  |
+----------------------------------------------------------------------+

--
Unsubscribe mailto:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi