DiRT, otherwise known as Disaster Recovery Testing, is Google’s attempt to break the web.
More specifically this team’s mission is to provide (and actually simulate) worst case scenarios to Google’s sprawling worldwide infrastructure, taking entire live data centers offline without notice. The end goal is to find weaknesses in their own disaster recovery plans and improve upon them.
The scenarios are extreme. They range from meteorite strikes, alien invasion, earthquakes, and nuclear war. But simpler scenarios such as fire, tsunami, and theft can be disastrous as well.
What does your disaster plan entail? Of course you must have the basics of food water and shelter, but what about your tech? Is it backed up? Is it backed up offsite? Have you ever tried to recover any data from backup and test to see if it actually works?
Knowing that your backups are functioning is just the first step to disaster recovery. What about the software you use? Do you know where to obtain it again? Was it on a disk or was it downloaded? If downloadable, do you have the credentials (usually a username and password for paid software) to access the site that has the download? Do you have a copy of license keys needed to reinstall software? Is all of this information located somewhere offsite so it can be quickly and easily accessed in the event of a disaster?
Now we will look at hardware. For most home users this entails 1-3 computers, some mobile devices and a printer. For businesses add managed switches, network printers, servers, patch panels, special configuration of hardware firewalls and routers and phone systems. Do you have a list of where to obtain all of that hardware again as soon as possible to minimize your downtime?
Some industries (medical, financial, utilities) are regulated on the tech side to where they must have a written and tested disaster plan to ensure minimal downtime. It just makes sense for the rest of us to do the same. Hopefully we’ll never need it – but I sure feel better knowing it’s there in case we do.