question

It sounds like you need to look at "SQL Server AlwaysOn Availability Groups". Although be warned - you might not have the right edition of SQL Server to do that properly. See [MSDN overview of AO AGs][1]. If you don't have the right version, then your options are more limited - you could do something like [logshipping or mirroring][2], or even [replication][3], but those would require intervention to complete a failover. What these things ultimately boil down to, though, is how much data can you afford to lose (Recovery Point Objective, RPO), how long can you keep running without your database being available (Recovery Time Objective, RTO), and how deep are your (corporate) pockets? Oh, and the amount of data we're actually talking about... [1]: https://msdn.microsoft.com/en-us/library/ff877884(v=sql.110).aspx [2]: https://msdn.microsoft.com/en-us/library/ms190202(v=sql.110).aspx [3]: https://msdn.microsoft.com/en-us/library/ms151198(v=sql.110).aspx

If you decide to use SQL Server AlwaysOn Availability Groups (AG), I would strongly recommend getting some external help when setting things up. Getting things up and running is not that hard (well.. even that's not entirely true, there are many details you need to get right Before getting it to work), but there are more things to monitor when you implement high availability. One example being that the cluster functionality won't automatically give you a phone call and tell you your primary node has failed over to the secondary, or that the secondary node has a problem. Don't hire a consultant to set it up for you. Hire a consultant to hold your hand and talk you through setting things up yourself the first few times (I'm assuming you'll both need to and want to practice setting these things up a few times before doing it in your production environment), and to make sure you don't end up with a false sense of security. Oh, and by the way - you should perhaps also include AlwaysOn Failover Cluster Instance (FCI) as an option. It would mean sharing disks between the nodes in the cluster, whereas AG is a "share nothing"-technology. Last time I was involved in such decisions, we sat down for an hour and a half with a DBA-consultant. My sketched plan was to implement AG but we ended up implementing mirroring instead, because it is a lot simpler to setup and, perhaps more important, easier to maintain.