Disaster Recovery in Another Region

  • Comments posted to this topic are about the item Disaster Recovery in Another Region

  • I consider this a basic part of delivering first class service to our customers.  We use Log Shipping for SQL Server data, and Hyper-V VM Replication for many critical servers.  The 850 mile trip from Orlando to Reston is "just a hop".  You don't need a cloud provider for this, you just have to decide it's worth it and buy the bandwidth.  We clamp the VPN to 100 Mb to avoid high costs, and it stays about 30% busy.


    Student of SQL and Golf, Master of Neither

  • This could be an exception, but as a general rule, disk replication does not allow for data integrity for database without gaps in data integrity protection.  The first is that a DBMS is a system that uses several stores - data, logs, checkpoint information - to present a uniform view of data.  With disk replication, it is inevitable that there will be some gap between the replication of all of these files, which could result in a loss of data integrity or even the inability to access the database at all.  In addition, disk replication is at the physical level, and if there is an error in the originating block (which could have been cause by some flaw in a write - very rare, but not impossible), that error will be replicated to the target block.  Granted, these are fairly small windows of danger, but they exist, and can result in massive problems if they occur.

Viewing 3 posts - 1 through 2 (of 2 total)

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