Considering the high cost of IT downtime, disaster recovery planning is critical for every enterprise. According to a 2016 IHS report, downtime costs North American companies $700 billion a year. For a typical mid-size company, the average cost was around $1 million, while a large enterprise lost more than $60 million on average, IHS found.
Yet even with the stakes so high, companies can fall into common pitfalls when it comes to disaster recovery planning to mitigate the impact of service outages. GS Khalsa, senior technical marketing manager at VMware, said that he sees organizations making the same three mistakes over and over again.
1. Not having a DR plan
In Khalsa's opinion, by far the biggest mistake that companies make -- and one of the most common -- is failing to put together any sort of disaster recovery plan at all. He said that industry statistics indicate that up to 50% of organizations haven't done any DR planning.
That's unfortunate because preparing for a disaster doesn't have to be as complicated or as costly as most organizations assume. "It doesn't have to involve any purchases," Khalsa said in an interview. "It doesn't have to involve anything more than a discussion with the business that this is what our DR plan is."
Even if companies decide to do nothing more than restore from their latest nightly backup, they should at least write that plan down so that they know what to expect and what to do in case of an emergency, he added.
2. Not testing the DR plan
Coming up with a plan is just the first step. Organizations also need a way to test the plan. Unfortunately, in a traditional, non-virtualized data center, there isn't an easy, non-disruptive way to conduct a recovery test. As a result, most companies test "infrequently, if at all," Khalsa said.
He pointed out that having a virtualized environment eases testing. Organizations can copy their VMs and test their recovery processes on an isolated network. That way they can see how long recovery will take and find potential problems without interrupting ongoing operations.
3. Not understanding the complexity of DR
Organizations also sometimes underestimate how much work it takes to recover from a backup. Khalsa explained that some organizations expect to be able to do their restores manually, which really isn't feasible once you have more than about 10 or 20 VMs.
He noted that sometimes IT staff will write their own scripts to automate the recovery process, but even that can be problematic. "People forget that disasters don't just impact systems, they also potentially impact people," Khalsa said. The person who wrote the script may not be available to come into work following a disaster, which could hamper the recovery process.
Khalsa's No. 1 tip for organizations involved in DR planning is for IT to communicate clearly with the business. Management and executives need to understand the recovery point objective (RPO) and recovery time objective (RTO) options and make some decisions about the acceptable level of risk.
"More communication is better," Khalsa said.
Hear more about disaster recovery planning from GS Khalsa live and in person at Interop ITX, where he will present, "Disaster Recovery In The Virtualized Data Center." Register now for Interop ITX, May 15-19, in Las Vegas.