Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.tools -> Re: Replicating to different database

Re: Replicating to different database

From: Bob Schmitz <bschmitz_at_execpc.com>
Date: Fri, 2 Mar 2001 12:11:01 -0600
Message-ID: <SlRn6.3721$7e6.1470629@homer.alpha.net>

If your looking for a COTS (commercial-off-the-shelf-software) package, look into using Sybase's replication server because I don't think you can get what you want using Oracle Replication. The Sybase Replication server uses functions strings (which is a fancy way of saying that it formats the data for the target RDBMS). I have never done this myself but I'm pretty sure that it will work. It will be a lot of work by yourself. You may want to get a hold of Sybase and asked them if they have something already made.

Hope this helps.

-rws

"David Dawkins" <david_j_dawkins_hotmail_com_at_devnull.com> wrote in message news:97lk3q$j3g$1_at_soap.pipex.net...
> Hi,
>
> I'm looking for a way to replicate certain tables from an
> Oracle (v7.x/v8.x) database to a different type of database
> (not necessarily an SQL-based database).
>
> Initial experiments use triggers to create a transaction table,
> which is polled and consumed by an external application, and which
> then updates the remote database. Unfortunately, the impact of
> the triggers combined with the "reader" application slows down the
> transaction turnaround time in the master application way too much.
>
> My questions, then:
>
> 1. All I really want is an API that will supply me with a transaction
> stream; is there any product out there that already does this?
>
> 2. Can Oracle maintain a transaction log that can be polled and
> read by a reader application? The reader application will need access
> to the transaction type (action,table name) and the data (field names
> and values).
>
> 3. Just supposing that triggers and a reader application is the only
> way to do this, what is the optimal way to insert the transaction
> details into the transaction table, and what is the optimal way to
> read and clean-up the transaction table? "Optimal" to me would mean
> minimal delay on the master application's transaction, while maximising
> the transaction throughput rate in the reader.
>
> Thanks for any pointers; even if the appropriate reply is "hey sonny,
> you just asked for about $200,000,000.00 worth of consultancy there.". :)
>
> Regards,
> David Dawkins
> david_j_dawkins_hotmail_com_at_devnull.com
> Please fix my reply address!
>
>
>
>
>
Received on Fri Mar 02 2001 - 12:11:01 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US