Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Usenet -> c.d.o.server -> Re: 10g to RAC - what to do to avoid data duplication ?

Re: 10g to RAC - what to do to avoid data duplication ?

From: Niall Litchfield <>
Date: Sun, 29 Apr 2007 12:54:10 +0100
Message-ID: <>

Golan wrote:

>> Excuse Sybrand, I was not so clear in the main question:
>> We have a 9i database on a server, we decided to create a new one 10g on

> the
>> new server, recreating the main objects and transferring historical data
>> with a db link.
>> Now we want to make all possible actions, to make the new server

> "raccable"
>> without duplicating data, and with the less effort possible.
>> Asm can give us just balancing becouse redundancy is guaranted by the raid
>> configuration on the storage.
>> Massimo

> The 10g server does not exist yet, I have to choose everything.
> I want to make a Rac in the future (next year) using this server as a node
> of the RAC.

Sybrand's advice is still good. If you are planning to go RAC then it makes sense to choose a storage medium that is compatible with RAC from the outset. There are a number of these now - RAW and ASM have both already been mentioned, in addition there are a number of NFS based solutions that are compatible and most likely Oracle's own cluster file system will be available to you as well.

I would, however, strongly query whether moving to RAC will actually happen. It sounds to me like a possible plan for later, rather than a firm intent - I maybe misreading you. Choosing more expensive, or more complicated, storage because you may do something at a later date seems a little foolish to me. Similarly if you are going to go RAC for sure it  would likely make sense to do it at the start of the project, when you have the downtime and testing associated with an upgrade project anyway, rather than later when you will likely have to repeat most of the work, but on a now live system.

Niall Litchfield
Oracle DBA
Received on Sun Apr 29 2007 - 06:54:10 CDT

Original text of this message