August 7, 2017 at 2:33 am
In my existing environment there is a design of distributed AG is proposed, can you please help to let me know few my queries regarding this.
1) what is the benefit of distributed AG which can not be availed from traditional AG.
2) what is the high availability benefit which we can not leverage with traditional AG, if you can explain one scenario of fail over which we can not avail in traditional AG?
3) in which scenario the design of distributed AG is preferable?
thanks
August 7, 2017 at 8:59 am
I think a bit of Googling would go a long way here, but a Distributed Availability Group can be thought of as an Availability Group of Availability Groups. They can span different WSFCs, and those WSFCs can even be on different operating systems (so you could theoretically do an OS upgrade from 2012 R2 to 2016 without almost no downtime).
There's a video here that is quite good.
http://www.pass.org/24hours/2016/summitpreview/Sessions/Details.aspx?sid=53726
August 7, 2017 at 7:06 pm
Zeal-DBA - Monday, August 7, 2017 2:33 AMIn my existing environment there is a design of distributed AG is proposed, can you please help to let me know few my queries regarding this.1) what is the benefit of distributed AG which can not be availed from traditional AG.
2) what is the high availability benefit which we can not leverage with traditional AG, if you can explain one scenario of fail over which we can not avail in traditional AG?
3) in which scenario the design of distributed AG is preferable?
thanks
With a single Availability Group, all replication to all nodes comes from the Primary Replica in the AG. In a Distributed Availability Group, the Primary Replica of each separate AG replicates to the secondaries in its AG. The separate AGs may be in different clusters and/or different domains.
The benefits come from using DAGs in DR planning or moving readable secondaries to a second location: place a fully-functioning AG in a second datacenter or branch office (or even add a third AG in a third location), and have a scripted-and-prepared run book for failing over to it - even if that's simply a matter or executing the ALTER AVAILABILITY GROUP command to switch the Primary.
Site-to-site replication is between each AG's Primary Replicas only, limiting bandwidth, and the cross-domain option allows flexibility with how some orgs handle site domains.
DAGs aren't for high availability; they provide additional replication flexibility for DR.
Eddie Wuerch
MCM: SQL
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply