[rrd-developers] If I could save time as a double ...

Dave Plonka plonka at doit.wisc.edu
Mon Nov 19 23:11:30 MET 2001



... I'd patch it and send it to you.

Ahem, anyway...

For some recent work, I find myself being required to store time-series
data using an interval (or `step' in rrdtool parlance) that has a whole
and fractional part, e.g. 337.5 seconds which will yield 256 data
points each day.

For this work, it is important that the number of data points per day
is a power of two, where-as the canonical MRTG/Cricket/FlowScan five
minute interval yields 288 data points per day.

I'm exploring the possibility of using rrdtool to do this.
Some options I have are:

0) Use rrdtool unmodified, but use it in a kludged way by lying to it
   about the "current" time on updates.

   I'm thinking that I could set the `step' to 1 when creating the RRD
   file, and then `rationalize' each timestamp into interval numbers
   (1, 2, 3, 4, etc.), since the epoch, before running update.

   There are some difficulties here, not the least of which is that it
   may be confusing and also that "rrdtool graph" will label the
   horizontal axis without any knowledge of how I have re-mapped the
   timestamp values, so I'm not sure rrdtool is buying me much if I
   must graph the data externally.

1) Modify rrdtool to allow a double as the step value.

   Is it possible to maintain binary compatibility for the data file
   (finding some place to store four more bytes for the step) or would
   this require a new version of the binary file format, and very
   many source code changes including dump and restore?

Thoughts?

It may help to give you a hint of what the work is about.
The Processing time-series data using some signal analysis techniques
benefits from the number of intervals being a power of two because one
can keep dividing the period in half (evenly), looking for signals of
successively shorter wavelengths, and helping to identify diurnal
behavior that we know is in the network traffic.

Thanks,
Dave

-- 
plonka at doit.wisc.edu  http://net.doit.wisc.edu/~plonka  ARS:N9HZF  Madison, WI

--
Unsubscribe mailto:rrd-developers-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi



More information about the rrd-developers mailing list