Okay, I've downloaded and installed Oracle on a single, Linux,
development machine many times now. That's not what this question is
about.
This is a question about what is required to implement a mission
critical, Enterprise quality, no Klingons
not wearing pants allowed,
Oracle system, especially any such system using Oracle Parallel
Server, or Oracle Real Application Clusters, or any system involving
some distributed deployment of Oracle.
What are the tasks required of sysadmins, dbas, developers, and even
the dudes from finance in implementing a large, mission critical,
distributed Oracle application? And how much time is actually
required of those tasks? I am not concerned with data modeling by
itself, but would like to know about testing and tuning of the
application, as well as tuning of Oracle and testing of Oracle, and
especially various Oracle failover/failback capabilities.
I can think of several tasks:
- specifying/pricing/approval/ordering/paying for the software
- specifying/pricing/approval/ordering/paying for support
- . . .
- training for sysadmin & dba
- installation of software across a cluster
- determining disk layouts
- creation of databases
- . . .
- securing the system
- testing security
- testing backup
- testing system in degraded modes
- testing failover
- testing failback/recovery
- documenting the system
- testing the documentation
- . . .
- . . .
Thanks,
