[rrd-developers] reducing disk IO load (wish list)

[Tobias Oetiker]

Hi Kevin,

the structure is



HHHHHHHHHHHHHHHHHHH
HHHHHHHHHHHHHHHHHHH
HHHHHHHHHHHHHHHHHHH
HHHHHHHHHHHHHHHHHHH

RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...
RRA1-DS1DS2DS3DS4DS5...

RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...
RRA2-DS1DS2DS3DS4DS5...


the only thing that is not interleaved which might be, is the 2h max
and average array ...

> My current solution is a perl wrapper around the RRDs lib to write to CSV
> files and a cron job to update the RRDs hourly or when grapher is run.
> reducing the caching requirement to 110Meg to do the real time updates.
> I see this is similar to RRD Accelerator but uses files directly not via
> sockets. When is 1.3 due?

1.3 is due when it is done :-)

regarding the accellerator, I have not started working on it yet,
and I am wondering if maybe a mmaped file as communication
'platform' may be better siuted. it would have the advantage, that
it could be synced to disk, makeing it crash resistant ...

cheers
tobi
-- 
Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten
http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902