How often do you test the recoverability of your systems? StorageCraft's Greg Wyman says daily automated recovery testing is needed – and it needn't be hard or expensive.
Data is the most critical asset for all businesses, regardless of size or complexity.
If an organisation were to lose three months of data, including all debtors, it would certainly go out of business.
When disaster strikes, it is too late to worry if backups are recoverable. New Zealand market feedback suggests only a small number of organisations fully test the recoverability of all their servers occasionally – any many not at all.
Yet those that don’t perform a full and complete recovery test every night, run a serious risk. Backup is often likened to an insurance policy – a drain on budgets until an emergency occurs, then management wish they had invested in a premium solution which included fast recovery and daily testing options.
Thousands of people each year find that some, or even all, their data is non-recoverable.
These businesses need DART - daily automated recovery testing.
Their testing should be fully automated and require no manual involvement, whether the organisation has 400 virtual servers across 18 hosts and multiple locations, or has a single physical SQL, Exchange or Windows Essentials server.
Staff should begin each workday by checking a screen to be assured that their data, databases and servers are recoverable - because recoverability was tested overnight.
Automated and complementary Recovery solutions that leverage the latest DART technology, such as WatchDog Software, take next generation sector-based real-time backup, recovery and disaster recovery technologies and automatically test the recoverability of every server nightly without manual involvement.
Recovery testing should cost less than NZ$3 per day per server. When a prospective customer tells me “We have a great backup, it cost $xxx,xxx and does everything,” I ask: ‘Do you test recoverability every night?”
Usually they respond: ‘Of course not, it is too difficult and time-consuming’.
But if it is too hard to test recoverability, imagine what it will be like if and when a disaster strikes – ouch! So recovery testing should be automated and occur every day/night.
Disaster recovery with automated recovery testing can often be added as a complementary solution to existing backup products, enabling an existing product to run between 7pm and 6am, and the real-time disaster recovery solution between 7am and 6pm to deliver near-instant backups every 15 minutes and night recovery testing of all your servers.
As an experiment, tell the IT team they have to do an immediate fire drill to check how long it takes to get three critical servers back into full production. Starting now!
Once the colour returns to their faces, monitor how long it takes, how disruptive it is and then determine how much data would have been lost.
From this exercise, you will be able to determine what the impact would be to the business, your users and your customers if the event had been real.