[rrd-users] Re: About logging the REAL valuesn

BAARDA, Don don.baarda at baesystems.com
Wed Aug 23 04:17:40 MEST 2000


G'day again,

I'm sure this is well understood by all, but thought I'd clarify anyway.

> -----Original Message-----
> From:	George Dau [SMTP:gedau at isa.mim.com.au]
> Sent:	Wednesday, August 23, 2000 11:00 AM
> To:	'rrd-users at list.ee.ethz.ch'
> Subject:	[rrd-users] Re: About logging the REAL valuesn
> 
> 
> My issue is that I have a hard time explaining to managers that 0.4 of an
> error count is a valid item. They do not believe me, and I am still
> unconvinced too. I really do want the number of errors since the last
> sample, not some averaged, munged, figure. Tobi is right, I have the
> source
> code so I can change it to do what I want.
> 
	The thing about rrd is it stores entries for samples taken at
discrete periodic time steps. Because real-world samples are rarely taken at
the exact time the entry is for, it _adjusts_ the sample to its best guess
for what the sample would have been at the exact time for that entry. 

	There are two ways to record and accumulate almost-but-not-quite
periodic samples; adjust the samples to match the periodic entries, or
adjust the entry times to match the samples. RRD does the former, some
applications might prefer the latter. The beauty of adjusting the samples is
the database is simple and compact, containing only samples for implicit
times. Adjusting the time intervals would require storing times as well as
samples for each entry, not to mention the complications of accumulating and
unknowns.

	If your problem requires exact samples at not-quite-periodic
intervals, then you need something different to RRD. Usually you will find
that either; you actually do want estimated values at exact time intervals,
or your samples can be (at least treated as if they were) recorded at the
exact times. Also, these two solutions become more alike the shorter your
sample interval... ie tend to the same solution as step tends to zero.

	ABO

--
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



More information about the rrd-users mailing list