[rrd-users] Re: disabling interpolation?

[Tobias Oetiker]

Today you sent me mail regarding [rrd-users] Re: disabling interpolation?:

*> 
*> Paul C. Williamson wrote:
*> > 
*> > Not to add fuel to the fire, but...
*> > 
*> > There were probably 5 people in the room from 9:00 am until 9:47, then 1 walked out leaving 4.  Over that hours, there were an average of 4.76 people in the room.
*> > 
*> 
*> Right.  As I said: don't mess with the averages.  However, it is safe
*> to say the minimum amount of people in that room is 4, not 4.76 and
*> the maximum amount of people in that room was 5, not 4.76 .
*> It makes no sense to say that during this interval there were 4.76
*> people in the room *unless* you are talking about averages.
*> 
*> This is why I think it is good to have a data source type which is
*> valid during the whole interval, independent of the time in this
*> interval that the data was entered.  See my other mail about this; it
*> makes it clear that the proposed "REAL" DST cannot be combined with
*> averages.
*> 
*> If you alter your example so that it talks about five minutes (and the
*> times can be 9:00 to 9:04 + 9:04 to 9:05) you have exactly the problem
*> that most people suffer from (assuming the default step time of 5 mins).
*> 
*> Either you measure at 9:00 and 9:05 *exactly* or you need another way to
*> work around the problem.  If I enter "4" for the number of people in the
*> room at 9:04, and the number used to be 5, I cannot say that there was
*> a minimum of 4.80 people in the room...

the point is that if you want to have higher accuracy and this is
what this is basically about then you have to increasse the
sampling rate ... because otherwhise it will be completely by
chance that you catch the 4 people in the room at 9:04 while there
were 5 people in the room at all other one minute points between
9:00 and 9:05 ... So in order to do this properly you have to set
the sampling interval to 1 minute and measure the population every
60 seconds ... all the rest is complete randomness ... 

cheers
tobi

PS: Not that I would recommend it, but you could also set the
--step to 10 seconds and have an RRA as a MIN consolidation type
storing data every 30 base intervals ... with a minimal required
heartbeat of 600 seconds ... but as I said this is RANDOM stuff ...

-- 
 ______    __   _
/_  __/_  / /  (_) Oetiker, Timelord & SysMgr @ EE-Dept ETH-Zurich
 / // _ \/ _ \/ / TEL: +41(0)1-6325286  FAX:...1517  ICQ: 10419518 
/_/ \.__/_.__/_/ oetiker at ee.ethz.ch http://ee-staff.ethz.ch/~oetiker


--
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