[rrd-users] Re: RRDTOOL Create Configuration

Nuno Emanuel Rodrigues nuno.erodrigues at optimus.pt
Wed Nov 14 11:14:00 MET 2001


Hello again.
Thanks for your answer.
I have some comments to do about yor RRAS...
With RRA:MIN:0.5:1:525600 you are storing exact values (without any
consolidation) for around.....4 or 5 years....large range of time.

You are creating 3 different RRA's .....The question i have...are you
creating 3 different graphs?
What i mean is.. if you are creating just one graph, probably with AVERAGE
CF, and then showing max, min with GPRINT:vname : CF :MIN/MAX. , you will
have a max/min value not present in the graph... This was discussed before i
think...

One important question i would like to do :

Like you said: 

	"If the resolution of the RRA is higher than the
> resolution of the graph, the data in the RRA will be further consolidated
> according to the consolidation function ( CF) chosen".
> 
Of course ploting a yearly ghaph i will have more then 100k values (with 300
step) , and RRA resolution higher then graph resolution, so some
Consolidation will have to be done when the graphs are created...

So supposing i have 100k values (around 1 year), and i want to plot yearly
graph with 500x500 pixels , i will have a average of 200 values for each
pixel? 
If thing works this way...wont be very slow 500x (averages of 200 values)
???



> ----------
> From: 	estreigh at nexl.com[SMTP:estreigh at nexl.com]
> Sent: 	Wednesday, November 14, 2001 1:37 AM
> To: 	Nuno Emanuel Rodrigues
> Cc: 	rrd-users at list.ee.ethz.ch
> Subject: 	RE: [rrd-users] RRDTOOL Create Configuration
> 
> I was thinking exactly the same thing... and I went ahead and gave it a
> shot. When creating a database it is prepopulated with Unk data, therefore
> the size is immediately evident.
> 
> Creating a database with a single data set is unlikely for most and
> certainly out of the question for my purposes, so my test includes four
> data
> sets. Also, I want to show the maximum of maxima and minimum of minima in
> my
> graphs (thanks Alex ;-)) so I will use three RRAs, one for MAX, one for
> MIN
> and a third for AVERAGE.
> 
> rrdtool create c:\rrdtool\output\estreigh-lptp	\
> 	--step 300						\
> 	DS:procPercentUser:GAUGE:900:U:U		\
> 	DS:procPercentInter:GAUGE:900:U:U		\
> 	DS:procPercentProc:GAUGE:900:U:U		\
> 	DS:procPercentPriv:GAUGE:900:U:U		\
> 	RRA:MIN:0.5:1:525600				\
> 	RRA:MAX:0.5:1:525600				\
> 	RRA:AVERAGE:0.5:1:525600			
> 
> Disk space used = 49Mb
> 
> I have not yet attempted to run regular updates to this database and
> therefore I have not built any graphs from it. It took approximately 8
> seconds to build the database, which I thought was pretty fast. The
> updates
> do not concern me so much, but graph creation needs to be pretty quick to
> meet the demands of my customers. I'll let you know how it goes...
> 
> As for you question regarding the method of consolidation, from the manual
> I
> quoath,
> " The DEF call automatically chooses an RRA which contains CF consolidated
> data in a resolution appropriate for the size of the graph to be drawn.
> Ideally this means that one data point from the RRA should be represented
> by
> one pixel in the graph. If the resolution of the RRA is higher than the
> resolution of the graph, the data in the RRA will be further consolidated
> according to the consolidation function ( CF) chosen".
> 
> Regards,
> Erik Streight
> estreigh at nexl.com
> 
> 

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



More information about the rrd-users mailing list