You administer all the deployments of Microsoft SQL Server 2012 in your company. You have two
servers in the same data center that hosts your production database.
You need to ensure that the database remains available if a catastrophic server failure or a disk
failure occurs.
You also need to maintain transactional consistency of the data across both servers.
You need to achieve these goals without manual intervention.
Which configuration should you use?
A.
Two servers configured in a Windows Failover Cluster in the same data center
SQL Server configured as a clustered instance
B.
SQL Server that includes an application database configured to perform transactional replication
C.
Two servers configured in the same data center
A primary server configured to perform log-shipping every 10 minutes
A backup server configured as a warm standby
D.
Two servers configured in different data centers
SQL Server Availability Group configured in Synchronous-Commit Availability Mode
One server configured as an Active Secondary
E.
Two servers configured in the same data center
SQL Server Availability Group configured in Asynchronous-Commit Availability Mode
One server configured as an Active Secondary
F.
Two servers configured in different data centers
SQL Server Availability Group configured in Asynchronous-Commit Availability Mode
G.
SQL Server that includes an application database configured to perform snapshot replication
H.
Two servers configured on the same subnet
SQL Server Availability Group configured in Synchronous-Commit Availability Mode
Explanation:
http://msdn.microsoft.com/en-us/library/ff877931.aspx
It’s H, but the explanation is innane. Here’s the real explanation.
https://social.msdn.microsoft.com/Forums/sqlserver/en-US/ed3665b5-c5f2-4046-b004-75a5dedab657/choosing-between-sql-cluster-and-alwayson?forum=sqldocumentation
transactional consistency is not defined as the transactions are the same between the two servers, which it would be in a cluster, it’s that a transaction got distributed either with a mirror or a replication and the transaction was saved at the mirror or subscriber before it was committed. Clusters use raid for their drive redundancy, they dont’ need to juggle transactions. In real life, a cluster would work far better but due to an artifact in this question, it is not the right answer.
0
0
H
0
0