Re: Looking for a scalable solution for many DB links

From: Thomas Kellerer <spam_eater_at_gmx.net>
Date: Wed, 10 Dec 2003 09:09:37 +0100
Message-ID: <br6kc6$27ojo8$1@ID-13919.news.uni-berlin.de>

Paul Moore schrieb:

> Sybrand Bakker <gooiditweg_at_sybrandb.nospam.demon.nl> writes:
>
>>4) Consider using Oracle replication.
>
>
> I'm not sure I see how. Unless you *are* thinking that I am talking
> about an application with its data spread over 100 databases (in which
> case replicating the data into one database is a reasonable option -
> not as good as centralising the data in one database, mind :-)). But I
> can see how you might have read my original message as implying that I
> was using a "design" like that.
>
> Sorry for being unclear.
> Paul.

I second Sybrand's suggestion for using Oracle replication.

From what I know, if you can do a SELECT * FROM table_at_dblink then you can as well do a CREATE SNAPSHOT .... AS SELECT * FROM table_at_dblink.

If you enable for FAST REFRESH (i.e. enabling snapshot logs on the source DB's) then you will only transmitt changed data as well, which will probably speed up the whole thing too. And you can schedule everything when defining the refresh groups

Thomas Received on Wed Dec 10 2003 - 02:09:37 CST