[rrd-users] Input values normalization

[Simon Hobson]

> There is nothing wrong with, for instance, setting steps==3600 and CF==MAX 
> to remember the highest temperature reported during an hour, and to have 
> this as the RRA with finest granularity.
> 
> You could have two RRAs, one remembering MAX the other MIN, and then show 
> the temperature range seen suring any hour (e.g. a red area for the max 
> temperature seen, overlayed by a blue area for the minimum temperature 
> seen).
> 
> Not sure if cacti can do this, that's for someone else to answer, but 
> RRDtool can do it.

I'm fairly sure you can, but you'd need to do your own (non standard) data source setups. I suspect you might be better doing your own data collection as I'm not sure how well Cacti handles lots of high-frequency collection tasks - I have a fair number of RRDs where I collect the data externally but graph it in Cacti.

But the accuracy will still only be as good as the collection frequency (and it's relation to the rate of change of the measured value). If the measured value can (for example) rise sharply and drop back again in between samples then even the max function won't tell you anything about it.