[rrd-users] Re: Calculating average when aggregating

[Geert van der Ploeg]

> O.K., my way of averaging would be like:
>
> DS1 DS2   AGGREGATED
>
> Timestamp1:
> 20 NaN 20
> Timestamp2:
> 10 NaN 10
> Timestamp3:
> 20 10 30
>
> Sum of the three timestamps is 60 so the average should be 20

If you choose the right options for the averaging of the different
timestamps (so that the result of the averaging is the same as this table,
things should come right)

So, make sure (with a smart RPN) that averaging a 'certain number' with an
'unknown value', becomes 'certain number'.



> If the Print-Command behaves like the graphing it will look like this:
>
> DS1 DS2   AGGREGATED
>
> Timestamp1:
> 20 NaN NaN
> Timestamp2:
> 10 NaN NaN
> Timestamp3:
> 20 10 30
>
> There is only one valid timestmp, so the average will be 30.

Yup, this is indeed only "if the print-command behaves like the graphing".

But since you can control what will be in the averaging-CDEF, this will not
happen.


>
> Have you ever tried this ?

Nope, I didn't try anything of this particular problem.


Disclaimer:
Maybe it looks like I'm really sure about this problem. This is not the
case. I'm a bit hypothetically too.
Like I said: I haven't tried all of this.


Geert van der Ploeg

--
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