scalable metrics database

Heiko W.Rupp hrupp at redhat.com
Sun May 1 10:42:36 UTC 2011


Hey,

[ moved to rhq-devel ]

Am 30.04.2011 um 18:58 schrieb John Sanda:

> "OpenTSDB was written to address a common need: store, index and serve metrics collected from computer systems (network gear, operating systems, applications) at a large scale, and make this data easily accessible and graphable."

Sounds interesting.

I have some time ago started a wiki page http://rhq-project.org/display/RHQ/Ideas+about+Caching
and so on, which also talks about metrics.
Management summary: RAM is so cheap, lets put the stuff into Infinispan.

That page also talks about other areas, where caching stuff in memory will help us a lot with our
current performance issues.

Michael: we have a lot of very expensive recursive queries in our code, that take a huge amount of cpu
on Oracle - even if the tables  (especially resources) fits completely in memory. Those tables warrant
re-structuring and even pre-calculation of some queries.

  Heiko



More information about the rhq-devel mailing list