On 2012-11-06, at 7:22 PM, Steve Shipway <[email protected]> wrote:
> Simon said: >> I'd suggest you want a DS for heating which is either 0 (off) or 1 >> (on), and update this as often as you can. And have a separate DS for >> cooling, again 0 or 1. > > I strongly agree; otherwise this makes higher-granularity graphs very > awkward, unless there is a 1cdp=1dp RRA that goes on forever to avoid > averaging, which would have other problems, of course. > > If you cannot do the calculation before storage, then maybe define two > calculated DSs that that the status DS (0,1,2) and convert to on calculated > DS that is heating 0/1 and one that is cooling 0/1 ? Once you have this, > then it is easy to get percentage use time -- just create an RRA that is > 1cdp=1day and the average of the heating DS x100 is the percentage time the > heating is on that day! Much simpler to do all the necessary calculations. > > Steve > Thanks Steve and Simon, I'll redo the DS structure with a separate DS for heating and a separate DS for cooling. Obviously, I didn't think clearly about storing 0, 1, 2 as the mode values straight in the DS. And to answer Simon's query, it is indeed a variable-speed, modulating, variable capacity (need more buzzwords here) furnace. But, alas, the website JSON data won't tell me what capacity it is running as. (Or I haven't looked hard enough at the correct places). All I get is the state of operation. Cheers, -- Haroon Rafique <[email protected]> Information Security & Enterprise Architecture, University of Toronto _______________________________________________ rrd-users mailing list [email protected] https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users
