[Chicago-talk] Graphing (plotting) time-series data?

Warren Smith warren at warrenandrachel.com
Thu Apr 14 09:23:36 PDT 2005


RRDTOOL (http://people.ee.ethz.ch/~oetiker/webtools/rrdtool/)

It's not perl-ish, but it *is* great for storing and graphing time-based
strip charts, especially "gappy" ones (Used in MRTG). It can highlight
parts of the chart where there are gaps in the data or just fill it in
with averages.

However, there is an *extremely* large learning curve. RRDTOOL is
difficult to get working, but the end result is a database that you can
generate graphs off of extremely fast.

Just my 2 centavos.

-Warren


On Wed, 2005-04-13 at 19:35 -0500, Greg Fast wrote:
> On Wed, 13 Apr 2005 17:48:16 -0500, "Wiggins d'Anconia" <wiggins at danconia.org> wrote:
> > Greg Fast wrote:
> > > On Wed, 13 Apr 2005 17:00:55 -0500, "Wiggins d'Anconia" <wiggins at danconia.org> wrote:
> > > 
> > >>GD::Graph can graph just about any data set just about anyway, 
> > >>especially when combined with ImageMagick, but again you would have to 
> > >>handle the time series data structure yourself and supply it correctly 
> > >>to the module. If you haven't checked it out you might want to.
> > > 
> > > 
> > > It's not just handling the data bit, but the labelling bit.  The data
> > > isn't going to be gappy, and generating dated axis labels by hand is
> > > something I'd rather not do.
> > > 
> > 
> > Can you give us an idea of the data?  I am not sure why you would need 
> > to generate axis labels by hand, especially if the data *isn't* gappy...
> 
> Oh, whoops, that should have been "*is* going to be gappy" (that'll
> teach me to fire mail off as I run off to catch the train...)
> 
> The data (this week) is basically a set of {start: timestamp, elapsed:
> milliseconds, success: boolean} tuples.  I've got a couple of hundred
> points spanning a couple of months, with various gaps (where I had
> complete connection failures, bugs, etc).  Ideally, I'd like to flag
> the points where success=false with a red dot or somesuch (otherwise,
> I'll drop them from the data set or plot them with a second line).
> 
> --
> Greg Fast
> http://cken.chi.groogroo.com/~gdf/
> _______________________________________________
> Chicago-talk mailing list
> Chicago-talk at pm.org
> http://mail.pm.org/mailman/listinfo/chicago-talk
> 



More information about the Chicago-talk mailing list