Distributed datastores ( was Re: CLI on server part 2 )

Heiko W.Rupp hrupp at redhat.com
Tue Jan 11 12:39:35 UTC 2011


Am 11.01.2011 um 10:56 schrieb Lukas Krejci:

> I very much understand your concerns about user-friendliness of setting stuff 
> up, but at the same time I'm becoming more and more convinced that we do need 

Speaking of setup: in a single server case, there would be need for a
distributed file system.

While I see that just a few scripts will have no impact on the DB, we should
in general consider alternatives to the DB when needing to distribute stuff
to more than one server. Setting up one more RHQ server to offload CPU
is still cheaper in setup and maintenance (and often licenses), than extending 
the DB box + storage array + ...
There are cases where strong consistency is needed and here a relational DB
is the clear winner. At other times, "eventually consistent" is good enough.

  Heiko


-- 
Reg. Adresse: Red Hat GmbH, Otto-Hahn-Strasse 20, 85609 Dornach bei München
Handelsregister: Amtsgericht München HRB 153243
Geschaeftsführer: Brendan Lane, Charlie Peters, Michael Cunningham, Charles Cachera



More information about the rhq-devel mailing list