[rrd-users] Creating a graph averaged over a week
Steve Shipway
s.shipway at auckland.ac.nz
Thu Nov 18 09:33:21 CET 2010
This sounds like the perfect application for the Holt-Winters analysis functions in RRDTool.
If you have RRD version 1.3 or (preferably) 1.4, then you can add Holt-Winters RRAs to the RRD file. These can be set up with a 1-week period, and will then give you not only the predicted value that you want, but also a confidence band and even a method of highlighting deviations from the expected values.
The documentation on H-W is fairly brief, and its a difficult one to understand, but the defaults and example parameters work fairly well.
Steve
Steve Shipway
University of Auckland ITS
UNIX Systems Design Lead
s.shipway at auckland.ac.nz<mailto:s.shipway at auckland.ac.nz>
Ph: +64 9 373 7599 ext 86487
________________________________
From: rrd-users-bounces+s.shipway=auckland.ac.nz at lists.oetiker.ch [rrd-users-bounces+s.shipway=auckland.ac.nz at lists.oetiker.ch] on behalf of krad [kraduk at gmail.com]
Sent: Thursday, 18 November 2010 12:46 p.m.
To: rrd-users at lists.oetiker.ch
Subject: [rrd-users] Creating a graph averaged over a week
Hi,
Ok background first, as this isnt as obvious as it sounds.
I'm logging a data feed of my household power consumption. I have got graphs working fine for the standard fair daily,weekly monthly etc. However they are bit spiky. Therefore what I would like to do is generate one for a generic monday, tuesday etc. By this a mean is keep a high resolution of data over say a 28 day cycle (maybe more). Then generate a day graph that plots the average value for x time in a day.
ie
add up all the data points for say 09:30:00 on a monday and / 4 and then plot that point and so on for all of the 4 monday data points.
I would also like a generic weekday, weekend, and week graphs
Is this something rrdtool can do? At the moment I cant see how I could as all the x axis values would be different as they are time based
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20101118/b9ad107f/attachment.htm
More information about the rrd-users
mailing list