We continue our look at optimizing DR KPIs, the benchmarks you use to measure how ready you are for a disaster.

In the previous part, on RPO, we covered ways to improve the recovery point objective. This part looks at strategies for optimizing the recovery time objective (RTO). Here is what we recommend.

1. Document everything

To work proactively, document every aspect of your system and of your restore and recovery process. Your engineers should not have to work out what to do in the middle of an incident. They should be able to simply copy and paste the commands they need from the DR procedure document.

2. Set up monitoring and have a support team

How can you start a recovery if you don't know that something has failed?
How can you start a recovery if you have no technical people to carry it out?

As we have noted before, monitoring and support are essential foundations of any recovery process. Make sure you have a system that keeps track of everything, and a team or a person who can do the actual work.

3. Use replication and re-synchronization technologies

This is the same advice as in the previous part, on RPO. If you can simply switch over to working backup infrastructure, you don't have to spend time restoring and recovering from a backup.

4. Automate

Automation is one of the best ways to improve operational efficiency and support business continuity. Identify every process that can be automated with a tool.

Use Infrastructure as Code and configuration management tools wherever feasible. They sharply reduce the risk of human error and speed up your operations.

Implemented the right way, automation can improve RTO by up to 50% and streamline your DR workflows.

Where it makes sense, automate the switchover as well and trigger it from system metrics. Smart automation outperforms people at many such tasks.

5. Test your BC/DR solutions regularly

This is the most important point of all. Make sure your business continuity and disaster recovery systems stay in working order throughout the year. We recommend testing them at least twice a year, and ideally every quarter. If something goes wrong while you are executing the DR plan, your RTO will grow significantly. Until a solution has been tested, you can't be sure it works.