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: how to impelement distributed database in oracle

Re: how to impelement distributed database in oracle

From: Mark D Powell <Mark.Powell_at_eds.com>
Date: 2 Feb 2005 08:02:59 -0800
Message-ID: <1107360179.525518.137050@f14g2000cwb.googlegroups.com>


For those unfamilar with the term "distributed database" it can in general be defined as a logical database design implemented using two or more physically distinct databases. Such a design requires the coordination of backup and recovery operations for all the physical databases involved.

In the case of Oracle database links are used to query and send DML operations between the physical databases. In the case of a three node setup: A, B, and C if it became necessary to recover database B to a point in time prior to the time that A and C are at then all three database would need to be recovered to that point in time. Otherwise the application data would be logically inconsistent between the three physical databases used to house the logical database.

Just creating a link and issuing a few distributed transactions does not mean you have a true distributed database. You are just using some of the distributed features. Prior to the introduction of RAC (previously called OPS) if you outgrew your machines and just buying a bigger one was not a solution you resorted to designing a distributed database.

HTH -- Mark D Powell -- Received on Wed Feb 02 2005 - 10:02:59 CST

Original text of this message

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