AlwaysOn or Logshipping for DR

  • HI Folks,

    We are planning to setup DR for SQL Server in a different physical location. Currently there is AlwaysOn setup for high availability for this db.
    Network bandwidth is slow from Prod to DR at this point. Which is the preferred option to setup for DR database, AlwaysOn or Logshipping?

    Please share your thoughts. Thanks!

  • Your RPO/RTO should determine the method.  
    Also, define "slow bandwidth", and what you used to determine that it is slow.  That may be perfectly adequate for your workload.  If you have a system that has relatively small changes to the data, then this slow bandwidth may be adequate for your needs.  

    How fast can the log backups be run, copied, and restored to the DR server?  As an example, if you are performing log backups every X minutes, and they are large, then the time to copy and restore needs to be factored into the RPO.  You need to guard against the next log backup completing, and the previous one has not been restored yet.  That adds even more time to the RTO.  

    AlwaysOn may provide a lower RPO. I would suspect that the latency would be lower, again, depending upon the volume of data.

    Michael L John
    If you assassinate a DBA, would you pull a trigger?
    To properly post on a forum:
    http://www.sqlservercentral.com/articles/61537/

  • SQL!$@w$0ME - Wednesday, November 1, 2017 9:42 AM

    HI Folks,

    We are planning to setup DR for SQL Server in a different physical location. Currently there is AlwaysOn setup for high availability for this db.
    Network bandwidth is slow from Prod to DR at this point. Which is the preferred option to setup for DR database, AlwaysOn or Logshipping?

    Please share your thoughts. Thanks!

    There is no right or wrong answer that anyone can give you - it really depends on your needs, especially those related to RPO and RTO. There are other considerations such what are the defined failures are you protecting against - under what circumstances would the disaster recovery plan be implemented, do you need automatic failover, are there bandwidth/infrastructure issues to take into consideration, are there cost limitations or considerations. And undoubtedly more that those.
    You probably need to think through all of that and then consider what direction you should go. I'd probably search some of the different articles that compare the different options, advantages/disadvantages and try to see what comes closest to meeting the business needs.

    Sue

  • here's an interesting article originally from the SQL Server Customer Advisory Team about this topic:
    https://blogs.msdn.microsoft.com/sqlcat/2013/11/20/sql-server-2012-alwayson-high-availability-and-disaster-recovery-design-patterns/

  • Michael L John - Wednesday, November 1, 2017 10:18 AM

    Your RPO/RTO should determine the method.  
    Also, define "slow bandwidth", and what you used to determine that it is slow.  That may be perfectly adequate for your workload.  If you have a system that has relatively small changes to the data, then this slow bandwidth may be adequate for your needs.  

    How fast can the log backups be run, copied, and restored to the DR server?  As an example, if you are performing log backups every X minutes, and they are large, then the time to copy and restore needs to be factored into the RPO.  You need to guard against the next log backup completing, and the previous one has not been restored yet.  That adds even more time to the RTO.  

    AlwaysOn may provide a lower RPO. I would suspect that the latency would be lower, again, depending upon the volume of data.

    Thanks Mike.

  • Sue_H - Wednesday, November 1, 2017 10:42 AM

    SQL!$@w$0ME - Wednesday, November 1, 2017 9:42 AM

    HI Folks,

    We are planning to setup DR for SQL Server in a different physical location. Currently there is AlwaysOn setup for high availability for this db.
    Network bandwidth is slow from Prod to DR at this point. Which is the preferred option to setup for DR database, AlwaysOn or Logshipping?

    Please share your thoughts. Thanks!

    There is no right or wrong answer that anyone can give you - it really depends on your needs, especially those related to RPO and RTO. There are other considerations such what are the defined failures are you protecting against - under what circumstances would the disaster recovery plan be implemented, do you need automatic failover, are there bandwidth/infrastructure issues to take into consideration, are there cost limitations or considerations. And undoubtedly more that those.
    You probably need to think through all of that and then consider what direction you should go. I'd probably search some of the different articles that compare the different options, advantages/disadvantages and try to see what comes closest to meeting the business needs.

    Sue

    Thanks Sue!

  • Chris Harshman - Wednesday, November 1, 2017 10:51 AM

    here's an interesting article originally from the SQL Server Customer Advisory Team about this topic:
    https://blogs.msdn.microsoft.com/sqlcat/2013/11/20/sql-server-2012-alwayson-high-availability-and-disaster-recovery-design-patterns/

    Thanks Chris.

Viewing 7 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic. Login to reply