locked
DB High Availability Design RRS feed

  • Question

  • Hello,

    A task to implement a high availability DB system has been put in our plate and it would be a first for a big project like this. It involves running a database in clustered Windows 2012 server environment and at the same time replicating the cluster offsite. We will be using 2 physical production servers for each site's cluster, and will run SQL server 2012. Here are some of questions to get things going:

    On each cluster, what's the best way for the storage? Do we really need a SAN or can we allocate a separate volume on each of the servers and have the database mirroring happening on the background?

    On each site, what's the best way for replicating the data from the main to the backup site? We have an 8Mbps MPLS connection between the 2 sites and which we can upgrade up to 1Gbps if necessary.

    Thanks in advance for your help.

    Thursday, October 31, 2013 2:17 AM

Answers

All replies

  • If you using SQL Server 2012 EE consider this

    https://www.microsoftvirtualacademy.com/tracks/mission-critical-confidence-using-microsoft-sql-server-2012

    http://www.sqlservercentral.com/articles/FCI/92196/


    Best Regards,Uri Dimant SQL Server MVP, http://sqlblog.com/blogs/uri_dimant/

    MS SQL optimization: MS SQL Development and Optimization
    MS SQL Consulting: Large scale of database and data cleansing
    Remote DBA Services: Improves MS SQL Database Performance
    SQL Server Integration Services: Business Intelligence

    • Proposed as answer by Shanky_621MVP Saturday, November 2, 2013 8:54 AM
    • Marked as answer by Fanny Liu Tuesday, November 5, 2013 8:53 AM
    Thursday, October 31, 2013 8:23 AM
  • Thank you Uri!

    I've been reading some materials related to my issue and I think your links will be very helpful. It's gonna be a busy weekend going through them.

    Regards,

    Trev

    Thursday, October 31, 2013 10:51 PM
  • First establish *exactly* what you and your business are wanting to achieve in terms of your availability and recovery before deciding upon a strategy. I suspect that you might want to consider AlwaysOn Availability Groups if you are going to use those versions. 

    If you go with a cluster, then you must use SAN storage via HBAs (or via iSCSI targets) and use as shared storage. You can of course mix various HADR strategies but doing so needs proper planning and a good working design knowledge.


    Regards,
    Mark Broadbent.

    Contact me through (twitter|blog|SQLCloud)

    Please click "Propose As Answer" if a post solves your problem
    or "Vote As Helpful" if a post has been useful to you
    Watch my sessions at the PASS Summit 2012 and SQLBits

    Thursday, November 14, 2013 11:07 AM