Data backup, long the forgotten sister of data center operations, has grown in prominence as data becomes increasingly central to business competitiveness and survival. It’s no longer enough to rely on a key employee bringing a backup tape home every night. Data center managers must look at backup in a broader context. Finding another way to describe it might be a reasonable place to start.

“We should be focused on data recovery, and more important than that, we should be focused on business recovery,” says Kelly Lipp, CTO of STORServer (www.storserver.com). “More than just backup, we need to ask ourselves how we get back to doing what it is we’re supposed to be doing in the event of a failure and how we do that quickly.”

Manage The Human Point Of Failure

Traditional data backup processes simply can’t keep up with today’s business needs. The human factor is often a critical point of failure.

“Take the simple example of nightly backups,” says Avocent CTO Ben Grimes (www.avocent.com). “If for some reason the person who’s been assigned to that task has to go home early one night for an emergency, the backup doesn’t happen. There’s no automated process in place to ensure your policy is enforced.”

Overreliance on manual processes opens up exposure on a number of levels, including disaster recovery and compliance. It also consumes resources inefficiently, which carries an opportunity cost, as well.

“Repeated studies indicate that the biggest cause of lost data or system downtime is human error,” says Matthew Parker, a project manager with Stantec (www.stantec.com). “No matter how well facility backup systems might be designed, we can’t eliminate human error. Properly maintained data backups can ensure that if support equipment fails to operate as intended, recovery times will be minimized.”

Plan For The Worst

Greater use of automation also frees IT resources to add more value in other areas. "Most people simply have difficulty accomplishing any kind of backup and struggle mightily just with that,” says Lipp. “Because they're so involved in these tactical actions, they don’t have any time to think about what they would do in order to recover. This near-sighted view of backup leaves IT organizations unable to focus on strategic planning, which would add value to—and reduce risk for—the rest of the business.”

Better strategic planning helps IT more thoroughly understand the business context of data backup. Kevin Watkins, vice president of tri-state engineering with Presidio Networked Solutions (www.presidio.com), says that before investing in improvements to backup capability, companies must first understand how backups can fail:

• The backup exceeds the allowable window, resulting in an incomplete backup.

• The backup encounters a problem, resulting in a corrupt or incomplete backup.

• Remote sites or servers remain outside of the backup process and are not properly backed up.



Any one of these can leave the organization without the ability to restore in the event of a disaster. Even when backups work as indicated, they may not be enough to effectively support business operations.

“During a recent upgrade, a client unexpectedly needed to restore from tape due to an unforeseen issue during the upgrade,” says Watkins. “The tape was mounted but had not been verified. As a result, the customer had to go to the last good tape, which turned out to be 30 days old, and rebuild the interim data by hand,” a process that took an additional two weeks and 100 person-hours of effort to complete.

Re-evaluate Your Backup Process

Situations like this reinforce the value of shifting from reactive to proactive backup.

“You have to admit you have a problem,” says Lipp, “that patching things up every six or 12 months isn’t working and it’s now time to re-evaluate your technology. It might mean you have to start from scratch, but if getting off the backup bandwagon means implementing a system that works once and for all and is absolutely scalable for your needs, then that’s how you’ll move forward.”

Once you grab the reins and implement more effective backup solutions and processes, make sure your operational processes keep pace.

“You have to measure and monitor the backup environment to make sure it’s up and running,” says Grimes. “I’ve seen so many cases where companies were able to back up but not restore, where they’ve tested their replication but never tested the restore side. These should be built into your data center’s day-to-day management processes.”

Make The Business Case, Carefully

Convincing decision makers to fund a more robust backup capability is always challenging because its benefits aren’t always easy to see. And larger organizations with extensive data centers packed with the latest hardware are just as vulnerable to backup weaknesses.

“This capability seems to vary more by the maturity of the IT organization than by its size,” says Grimes. “I’ve seen a lot of companies that have grown dramatically over the years. And to fuel that growth, they need more and more computing power and ended up throwing more equipment into the data center. But they didn’t build mature IT processes around them, so they ended up with large data centers and a relatively immature ability to manage them.”

Sometimes, the lessons are hard-won. “Often, companies have to get really mature really quickly after they’ve had their first failure,” he says. “Of course, by then it’s too late. They realize, belatedly, that it’s a form of insurance.”

He recommends approaching decision makers before any loss occurs to frame it in this context. “Nobody buys airbags for their cars—they just come as part of the package—but you sure want them to work when there’s an accident,” says Grimes. “You have to explain it to your decision makers in similar terms. If their CRM data, for example, is what drives the business and defines its value, have them quantify and qualify their exposure if suddenly tomorrow they don’t have it anymore. Talk about it in terms of the fear of loss.”

by Carmi Levy