Re: Hardware clustering solutions

I am sorry if i misunderstand something, but:

requirements don't specify, they require. If they do specify, they go beyond their scope.

After investigating requirements about (a.o.) maximum downtime/maximum recovery time and allowed amount of data loss one can start to investigate what would help to meet these requirements. That might end up into specifications.

Apart from that, clustering will save your uptime (maybe), but not your database. A cluster has 1 (one) database. A standby database solves this SPOF.
The uptime might be affected negatively by the added complexity. Complexity and availability are enemies. KISS.

Best regards,

Carel-Jan Engel

===
If you think education is expensive, try ignorance. (Derek Bok) ===

On Wed, 2005-11-02 at 19:42 -0500, Glenn Stauffer wrote:

> We're looking at a new server configuration and disaster recovery
> requirements specify that we have duplicate servers located in
> separate buildings on our campus. Currently, we're running an Oracle
> standby database to backup our production system. Given that we will
> have two of whichever servers we go with next, I'm exploring the cost
> and benefits of clustering solutions from HP, IBM, and Sun. We're
> thinking of consolidating our Oracle application and database servers
> on the two new servers and using the native clustering solution to
> provide automatic fail-over for these servers. I know someone using
> IBM's HACMP software who is very pleased and that sounds like a good
> possibility, but we're keeping our server decision open at the moment,
> so I want to look at what each of these vendors offers. Anyone out
> there with experience on any of these clustering solutions in an
> Oracle environment? Good experiences? Bad experiences?
>
> Thanks,
>
> Glenn Stauffer

--
http://www.freelists.org/webpage/oracle-l