5 Replies Latest reply: Jun 9, 2014 11:00 AM by user10955371 RSS

    Ideal DR solutions




      Can anybody tell what's the ideal DR solution between Solaris x86(Production) to Solaris SPARC (DR) with

      We can move data across platforms with TTS or expdp/impdp(exp/imp). But this is not going to be a complete DR as many permission/data type conversion we need to consider.


      But, is it possible? Any drawbacks and things to consider before implementing this.





        • 1. Re: Ideal DR solutions



          Data Guard rules. See:


          Data Guard Support for Heterogeneous Primary and Logical Standbys in Same Data Guard Configuration (Doc ID 1085687.1)


          Your best bet would be to match the OS on the DR side.


          Import and Export is not DR, it assures you have something but you will almost always lose data. Data Guard gives you a complete DR system, Failover, Switchover, backup on DR side etc.


          Best Regards



          • 2. Re: Ideal DR solutions

            I am agree with mseberg


            Also check if you can configure storage replication between both servers, it works well for DRP solutions. I´m not sure because you have different OS.



            • 3. Re: Ideal DR solutions

              > what's the ideal DR solution between Solaris x86(Production) to Solaris SPARC (DR) with


              Well that statement right there is in direct opposition to itself.


              If this were an "ideal" DR solution, the DR platforms would mirror the production platforms. You can't have an ideal DR solution and be on different OS platforms.


              Before you design your DR solution, it would be best if you could get your DR requirements documented. Your DR site is your safety net should the corporate primary platform be unavailable for some reason. So one would need to answer questions like:


              • How much data loss is acceptable?
              • How much time can we be down before we have systems up and running at the DR site?
              • What disasters are we trying to mitigate?
                • Natural disasters, i.e. fire/flood in the primary data center?
                • Man-made disasters, i.e. accidentally dropping a table.


              In my opinion, trying to design a DR solution without a thorough understanding of what we are trying to protect and how is the wrong approach. To me, this is similar to designing a backup solution before understanding the recovery requirements. I talked about that here:




              The whole purpose of a backup is to be able to recover from it. Similarly, the whole purpose of a DR solution is to be able to handle a disaster. It would be career suicide to design a DR solution that has data loss in it when management not only insisted on a zero data loss solution, but was also willing to pay for it!


              Last word on DR solutions....no matter what DR solution you design an implement, test, test, and test the solution! It was just 2 days ago that I participated in our twice-yearly DR test of our systems. Not only did we ensure that our DR implementation meets our requirements, we also ensured that any changes to our systems in the past 6 months did not negatively impact those requirements. We are better prepared to handle the disaster should that day become a reality.



              • 4. Re: Ideal DR solutions

                To setup DR site, You need to setup same OS and Oracle binary both sides. The best you can setup physical standby with Maximum Performance mode. If required you can set up multiple DR sites. Again you can try to use datguard broker to automatic failover in case primary is down.



                • 5. Re: Ideal DR solutions

                  Thank you all for your valuable inputs..user8914886 , BPeasland ,J.A.