V3 Summit: You can't do backup and recovery 'the old fashioned way' any more, warns Dell

Dell's Peter de Vente: Backup and recovery will have to change when organisations routinely hold hundreds of petabytes of data

Slow recovery from disasters or other downtime is handicapping organisations' ability to respond to unforeseen events, with less than one-quarter of organisations confident of being able to make a "full recovery to agreed timescales", according to research commissioned for the V3 Cloud & Infrastructure Summit.

Responding to the research, Peter de Vente, a systems consultant and business continuity specialist at systems and services giant Dell, suggested that organisations need to make sure that their disaster recovery plans include prioritising systems, databases and data so that IT staff can focus on getting the most valuable IT assets up-and-running first.

"It's a good question. How much time do you need for your restore? And you have to put that against how much data you need to restore. Then it becomes another story," said Vente.

Indeed, Vente suggested that organisations ought to look at how much raw data that they are storing right now and then scale up to take account of how much data they are likely to be packing in ten years time in order to build an enduring data recovery and back-up strategy.

"I always joke that in 10 years' time I won't have a job any more, because if you look at the pace of data growth and how we deal with data protection, it simply won't work any more when organisations have hundreds of petabytes of data. You can't do that in the old-fashioned way," Vente warned.

One of the challenges many organisations face when disaster strikes, continued Vente, is that staff supposedly working to bring up the same systems don't communicate and work together effectively.

"In a lot of companies, I see different admins - someone is responsible for data and data protection, and another person for the application. As long as they do back up, it works. At the point of restore, though, they need to work together. It's not that they can't, but that suddenly they need to talk to each other.

"They need to ask each other, 'how does this relate to our backups? How can I know what version is what?' There's sometimes a language difference, too - the database administrator wants 'this', but the back-up administrator only has 'that'," said Vente, who was speaking in a Summit panel debate entitled "Identifying the cost benefits of a strategic approach to enterprise backup and recover" alongside Andy Boura, senior information security architect at information company Thomson Reuters.

Boura pointed out that while building and maintaining a backup and recovery infrastructure does cost money, the cost in terms of lost business is even greater - as reflected in the research, conducted by Computing Research, among CIOs and other IT leaders.

"The average for a Fortune-1000 company would be $100,000 per hour. If you have a day's outage, that's an awful lot of revenue," said Boura.

The research also noted a big difference between the time IT leaders think it would take to recover data in the event of a disaster, and the time it might take to bring applications back online. "There's pressure to keep that [time] down, but I'm not surprised that it is expected to take longer to restore applications than data because there's a lot of additional complexity with applications," said Boura.

He continued: "You have got dependencies - potentially external dependencies - versioning; and, you have got the risk of partial or incomplete processing of data [when the application went down]. You may even have corruption of data that you have got to deal with and that can mean an awful lot of high-adrenaline work."

Restoring data, he added, is typically more straightforward than getting an application back up-and-running - provided backups are up-to-date and easily accessible.

Missed the V3 Cloud and Infrastructure Summit 2016? Tune in to the case studies and panel sessions - they're online now.