I have 5 database(Azure VM) and will be used as high volume transaction databases. I want to do reporting on these databases at peek times. Also to note that we may plan for Dataware house in future. I did some search and analysis(still in process) and come up with some options
1. Enable CDC on source databases and load these cdc data into new DB which can be used for reporting(without effecting main source db). This option will use ETL task.
2. use High availability where one node1 is primary and node2 as failover secondary and node 3 secondary for reporting.
3. Store data in cache(not sure but Azure catch may be) and latter used by reporting service
I have used mirroring,replication and cluster functionalities in past but not used high availability.
Can somebody help me on this. I know this forum is not on caching but if anyone has any idea then can share their views.
any suggestion on which one would be best?.