[rrd-users] Re: RRDtool Performance and number of RRA's

[Alex van den Bogaerdt]

On Fri, Nov 17, 2006 at 04:03:32PM -0000, Richard Burton wrote:

> I am running various tests to see what my maximum amount of updates per
> second my hardware can cope with and this appears to be greatly affected
> by the number of RRA's in a given RRD archive.
> 
> What appears to be happening is that increasing the number of RRA's from
> 2 to 6 has a dramatic effect on how many updates I can undertake per
> second. For example I started with
> 
> "RRA:LAST:0.5:1:525600",
> "RRA:MAX:0.5:1:525600",
> 
> and then moved to
> 
> "RRA:AVERAGE:0.5:1:129600",
> "RRA:AVERAGE:0.5:10:38880",	
> "RRA:AVERAGE:0.5:30:52560",		
> "RRA:MAX:0.5:1:129600",
> "RRA:MAX:0.5:10:38880",	
> "RRA:MAX:0.5:30:52560",	

This is not a simple change.  You have altered more than just
the amount of RRAs.

> My update speed has fallen by around 60%. Is this what one would expect?

RRDtool has to keep track of the data (going to go) in it for each RRA.
How it is going in (average, max, last) is also important, as is the
number of data sources.

More work to do, so more time needed.  I guestimate the amount of work
to be more than three times as much, so 60% off sounds nice actually.

Also think about memory requirements.  Perhaps the stats are influenced
by swapping as well.  This could happen if rrdtool, or another process(!),
needs memory and something else has to move from resident memory.

-- 
Alex van den Bogaerdt
http://www.vandenbogaerdt.nl/rrdtool/

--
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://lists.ee.ethz.ch/rrd-users
WebAdmin    http://lists.ee.ethz.ch/lsg2.cgi