Having a secure backup is great, but if you can’t recover your data in a timely manner you’re still going to suffer loss.
Most companies understand the importance of backing up their data, and just about every business has some kind of backup strategy in place. The problem, however, is that too often companies put so much emphasis on backing up data, including creating backups in multiple locations, that they neglect the other part of the equation — data recovery.
A global study conducted by Kroll Ontrack in 2016 involving 600 IT administrators provides further insights on this matter. While 57% of respondents had a backup solution in place, three quarters of that group admitted they were unable to restore all of their lost data following a server failure. What’s more, nearly one in four (23%) were unable to recover any of their data following a system failure.
Focus on Disaster Recovery
Instead of only focusing on the backup in BDR, companies are much better served turning their attention to the disaster recovery (DR) part of the equation. DR is focused on protecting a business from a server failure or other significant event, so it can maintain or quickly resume critical operations. A company’s disaster recovery plan depends on how much downtime it can accept. Some organizations may be okay with not having access to their data for four hours whereas others can only tolerate two to three minutes. When addressing these issues, two factors come into play:
- Recovery point objective (RPO): This metric deals with how much data loss is acceptable following a DR. For example, if a company only backs up its data once a day at the end of the work day, it’s possible that it could suffer a server crash near the end of the day, which would mean that it could potentially lose eight hours’ worth of data. If that’s not acceptable, it would need to perform backups more frequently throughout the work day.
- Recovery time objective (RTO): This metric entails the time it takes to recover your data and get up and running again. If a company stored all of its data in the cloud, for example, and the cloud app made data copies every five minutes, that would be a short RPO period. But, if it took three days to restore the data from the cloud (via the wide area network) that RTO would likely be outside most company’s acceptable parameters. For shorter RTOs, companies need a dedicated BDR appliances (or a network attached storage appliance) on-site, which are designed to recover large data sets in a matter of a few hours. For companies that require even shorter RTOs, virtualization technologies are needed.
Test Your Backups Regularly
The only way to ensure a backup set is good (i.e., it could be restored if you needed it), is to test your backups (or pay an IT solution provider to do it for you). Although the process will vary from organization to organization, the common thread in testing backups and restores is to perform a complete restoration of every last file to a clean system. Regular tests and test results should ensure your backup strategy is working.
It’s important to test restores against as many simulated hardware, software and service failures as are possible in the real world. To ensure all backup and restore instructions work as prescribed, you must re-test after updates that require change management and verification that existing backup media work with new backup hardware and software, take effect.
Using virtualization technologies is a great way to minimize the testing time and allow for more frequent tests.
When looking for the optimal BDR solution and planning their backup and recovery strategies, IT professionals should carefully consider their RPO and RTO needs and test their backups regularly to avoid any unpleasant surprises when the time comes to perform a restore. If there’s one thing that’s for certain – there will come a time that they’ll need to perform a data recovery.