[rrd-developers] Re: Implementing TOTAL

[Mark Mills]

> -----Original Message-----
> From: Alex van den Bogaerdt [mailto:alex at slot.hollandcasino.nl]
> Subject: Re: [rrd-developers] Re: Implementing TOTAL
> > > 1) average(all rates)*timespan(graph)
> > > 2) sum(for each defined interval:(rate*stepsize))
> > > The difference would be the unknown intervals.  In the first case
> > > unknown intervals are assumed to be at the average rate, in the
> > > second case unknown intervals are assumed to be zero.
> > 
> > Do both? TOTALSUM, TOTALAVG are really ugly suggestions for 
> names. =)
> 
> They're not clear to me and this is within context :)

Ouch. I did say they were ugly names. ;) In any case, TOTAL
should be "TOTAL" and leave unknown treated as zero. If you
want to make up data it isn't a TOTAL but an adjusted total
and your next problem will be what KIND of adjustments to
make. Some people might want an interpreted adjustment rather
than presuming a raw average for the unknowns.

If nothing else, maybe TOTAL and ADJUSTED would do?

Would thse work for TOTAL and ADJUSTED? (No dev box yet *sniff*)

	case VDEF_TOTAL: {
	    double sum=DNAN;
	    for (step=0;step<steps;step++) {
		if (finite(data[step*src->ds_cnt]) &&
!isnan(data[step*src->ds_cnt]))
		    sum += data[step*src->ds_cnt];
	    }
	    if (!isnan(sum)) {
		dst->vf.val  = sum*src->step;
		dst->vf.when = 0;	/* no time component */
	    } else {
		dst->vf.val  = DNAN;
		dst->vf.when = 0;
	    }
	    }
	    break;
	case VDEF_ADJUSTED: {
	    int cnt=0;
	    double sum=0.0;
	    for (step=0;step<steps;step++) {
		if (finite(data[step*src->ds_cnt]) &&
!isnan(data[step*src->ds_cnt])) {
		    sum += data[step*src->ds_cnt];
		    cnt ++;
		}
	    }
	    if (cnt) {
		dst->vf.val  = sum/cnt*steps*src->step;
		dst->vf.when = 0;	/* no time component */
	    } else {
		dst->vf.val  = DNAN;
		dst->vf.when = 0;
	    }
	    }
	    break;

Hey, since I'm cut-n-pasting your code around, why does AVERAGE
have step++ in it twice??? I took it outta mine because it looked
like a holdover from the while(){} loop style you had in much of
the other VDEF_ cases.

	case VDEF_AVERAGE: {
	    int cnt=0;
	    double sum=0.0;
	    for (step=0;step<steps;step++) { /* ++ here ? */
		if (finite(data[step*src->ds_cnt])) {
		    sum += data[step*src->ds_cnt];
		    cnt ++;
		}
		step++;   /* and ++ here too? */
	    }
	    if (cnt) {
		dst->vf.val  = sum/cnt;
		dst->vf.when = 0;	/* no time component */
	    } else {
		dst->vf.val  = DNAN;
		dst->vf.when = 0;
	    }
	    }
	    break;

> > Or alternately make TOTAL a raw sum and supply a COUNT and 
> STEPS so we
> > can calculate it manually?
> 
> We could have "preset" VDEFs such as timespan (number of 
> seconds graphed).
> This could then be multiplied by the calculated average and 
> thus provide
> the total as calculated using option 1 above.
>     VDEF:total=ds,AVERAGE,timespan,*
> This requires being able to use real RPN in VDEF commands, something
> that needs to be done anyway but it is currently not implemented.
> 
> Option 2 could be implemented by doing
>    VDEF:total=ds,TOTAL
> 
> Apart from calculating the total, is there any other need to know how
> many intervals were unknown or not?
> 
> Thanks for the input, I really appreciate it.
> -- 
>    __________________________________________________________________
>  / alex at slot.hollandcasino.nl                  
> alex at ergens.op.het.net \
> | work                                                        
>  private |
> | My employer is capable of speaking therefore I speak only 
> for myself |
> +-------------------------------------------------------------
> ---------+
> | Technical questions sent directly to me will be nuked. Use 
> the list. | 
> +-------------------------------------------------------------
> ---------+
> | http://faq.mrtg.org/                                        
>          |
> | http://rrdtool.eu.org  --> tutorial                         
>          |
> +-------------------------------------------------------------
> ---------+
> 

--
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