Hi,
We run our SQL 2016 environments in a 4 node setup. 2 node in primary DC use sync replication. 2 nodes in secondary DC using async replication. this meets out RTO/RPO needs. SQL is the backend for a html 5 web based product with multiple clients per 4 node setup.
In our current setup we have an availability group for each client - the Product contains 4 databases and also utilizes readable secondary's. we have the Availability groups spread across both nodes in the primary DC to balance the load.
We have an internal organizational change where we have a centralized DBA team. so our product is moving to there environment.
In their setup they don't use multiple Availability groups. The have 1 AG and all the DB's are added to this one AG. So they don't spread the load across both nodes in the primary DC.
What is the preferred setup, should you setup multiple AG's and spread the load. Or If you running in synchronous commit is the difference in load negligible so it doesn't make much difference having 1 AG on the primary node with all DB's in it?
I would be interested to hear peoples opinions on this and what you do?
Thanks.