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

Home -> Community -> Usenet -> c.d.o.server -> Re: Selective Replication

Re: Selective Replication

From: Sybrand Bakker <postbus_at_sybrandb.demon.nl>
Date: Wed, 16 Aug 2006 19:37:39 +0200
Message-ID: <erl6e2t5sc470704pcs38u258fbt6nen2c@4ax.com>


On 16 Aug 2006 09:17:27 -0700, "bernard (bernard_at_bosvark.com)" <bernard_at_bosvark.com> wrote:

>Scenario: Have two schemas (A and A_HIST) in the same database. A is
>used for currents (old data gets deleted) an A_HIST is used to capture
>data from A. Data never gets deleted from A_HIST. So A_HIST is currents
>and all the historic data too used for reporting.
>
>Current Implementation: Using Triggers on 20 tables in A to replicate
>insert and updates to A_HIST tables.
>
>Problem: I want to reduce the resource overhead (latches) of triggers
>and have flexibility of possibly have the A_HIST schema in a different
>database. I know database links on the triggers will provide this
>flexibility but all SQL and data changes are already captured in the
>redo and archive logs so Advanced Replication (using streams) should be
>faster, right?
>

Advanced Replication != Streams.
Advanced Replication uses *triggers*, Streams doesn't!

>Trotting though the documentation I can not determine if it is possible
>to only replicate INSERT, UPDATE and MERGE statements over the
>replicated schema using Advanced Replication, has anyone done this
>before? Had a quick search on asktom too, no luck.
>

Replication=Replication=Replication.
So
INSERT, UPDATE and DELETE will be replicated. Merge = INSERT + UPDATE

>Resources available: AIX 5.3, 10.2.0.1 (soon 10.2.0.2) and RAC
>
>I'm open for other solutions too.
>
>Thank you in advance
>
>Regards
>Bernard

--
Sybrand Bakker, Senior Oracle DBA
Received on Wed Aug 16 2006 - 12:37:39 CDT

Original text of this message

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