High availability keeps your SQL Server running when something fails. Disaster recovery gets you back after something bigger fails. They are different jobs, and most outages we see come from confusing the two, or from a backup nobody ever tested. The right setup starts with two numbers: how much data you can lose (RPO) and how long you can be down (RTO). Everything else follows from those.
A backup you have never restored is not a backup. It’s a hope. We test the restore, so you know.
HA and DR are not the same thing
High availability is about staying up. A second server takes over in seconds when the first one dies. Disaster recovery is about coming back. A copy in another site or region you can fail over to when the building, or the cloud region, goes. You usually want both. They solve different failures.
Your options, plainly
- Always On Availability Groups. The modern default. Database-level failover, readable secondaries, available on Standard and Enterprise with different limits.
- Failover Cluster Instances. Whole-instance failover on shared storage. Solid for lift-and-shift HA.
- Log shipping. Old, cheap, reliable. A good DR fallback when budget is tight.
- Backups to another region. Your last line. Test the restore or it does not count.
How to pick
- Write down your RPO and RTO per database. Not every database needs the same protection.
- Match the cheapest option that meets those numbers. Do not buy Enterprise HA for a database that can lose an hour.
- Test the failover and the restore. On a schedule. Before you need them.
Where it goes wrong
The common failures: a backup chain that was broken for months, a secondary that fell behind and nobody noticed, an RTO on paper that takes four hours in practice because no one had rehearsed it. We check all three.
Frequently asked questions
Do I need Always On, or are backups enough?
It depends on your RTO. If being down for a few hours is fine, tested backups may be enough. If you need to be back in minutes, you need a failover option like Always On.
Does HA protect me from a bad delete or ransomware?
No. HA copies the damage to the secondary instantly. Backups and point-in-time recovery protect you from that. You need both.
Can I run HA on Standard edition?
Yes, with limits. Standard supports basic availability groups: one database per group, no readable secondary. We work out whether that fits or whether Enterprise pays for itself.
Book a resilience review
Not sure your failover works? We will test your restore and your failover, and tell you the truth about your RTO. Book a resilience review.
Want to know if this is sitting in your estate? We run a read-only check and hand you a graded report in plain English.
Get your free health check