That disaster recovery plan, how do you get started with it? Well, that’s different for every organization. But an event like this underscores more than once the importance of having an action plan ready in the background. That’s why we’re happy to help guide you.
RTO AND RPO
To determine the best disaster recovery solution for your company, it’s important to map out your RTO and RPO.
- With RTO (Recovery Time Objective) you can express how long it can take at most for your entire setup to be back up and running after an issue. For a business-critical solution, this is often a matter of minutes, whereas a downtime of a few hours might be no problem for other organizations.
- With RPO (Recovery Point Objective) you can determine how much data you can afford to lose. In other words, how frequently you need to choose a recovery point (backup). For a busy webshop, this is an important indicator, since you obviously don’t want to miss many orders.
Once you’ve considered these two factors, you can work with your hosting partner to develop an ideal solution. Keep in mind that the lower the required RPO and RTO, the more expensive the solution will be. It’s also important to assess whether the cost is justified by the potential revenue lost due to downtime.
HUMAN FACTOR
It’s easy to define your RTO and RPO in an ‘ideal’ disaster scenario, but the human factor also plays a role. During downtime, data loss depends on the decisions made by the engineer working on a solution. If the engineer quickly decides (short RTO) to restore a full backup, data loss will occur (RPO). If the engineer chooses to avoid data loss and takes longer to find a solution, the RTO will increase.
SIZE OF YOUR DATA
The size of your critical data is crucial for selecting the right solution. If your data is too large, an offsite backup solution may take too long to rebuild in a new environment, exceeding the RTO. In that case, faster backup storage or higher network speeds are possible solutions. However, if you want to limit the RTO to just seconds or minutes, a hot standby or active-active failover in a second data center becomes necessary.
