There's nothing like a disaster to get people thinking about, well, disaster. If there can be any positive spin put on the Midwest flooding and Southwest wildfires, it's that they at least get the top brass of companies everywhere to give some thought to their own DR policies.
The trick is to make sure those concerns remain on the front burner during budget season, long after the actual disasters have faded from the headlines.
One of the ways to do that is by offering up a solid review of current plans, with particular attention paid to how some of the weaker links in the chain would have fared during the latest round of calamities.
Some of the latest thinking on building DR frameworks is to get beyond the notion that they are intended as temporary solutions designed simply to get you through a rough patch. Disaster recovery needs to become an integral piece of the overall data center environment, deserving of as much TLC as the server farm or storage array.
Chris Loeffler of Eaton Corp. argues that all locations that house IT equipment should be treated as a data center, with climate control, raised flooring and all the protections that the main center receives. All physical enclosures should have things like locked access, temperature management, power protection and distribution and full monitoring to prevent accidental, or intentional, damage. Small businesses should dispense with the notion that they don't need the same kind of protection as top firms.
Symantec's Dan Lamorena has outlined his five strategies for rethinking DR plans. He calls for increased monitoring and diagnosis, because sometimes it's hard to tell what's going on at a remote site, as well as increased use of automation, virtualization and remote backup to ensure that something in your enterprise is working somewhere. And it pays to test your system regularly.
Software AG's Bruce Beaman and Becky Albin say companies should approach disaster recovery the way an alcoholic approaches sobriety, with a seven-step program that starts out with the admission that they are, in fact, vulnerable. Once that is done, the process consists of assessing specific potential problems, inventorying infrastructure and IT assets, and determining the most effective way to maintain service level agreements. And again, it's a waste of time and money to initiate a DR program only to have it sit in mothballs until it's needed. You have to test, test, test.
Another key element in a successful DR program is the realization that it is not IT's responsibility alone, according to Dr. Vas Srinivasan at Sonasoft. All departments need to take careful stock of their data to decide what is critical for backup and what is superfluous. And since the importance of data changes over time, it must be continually reviewed for relevance. It's also a mistake to overwhelm your DR infrastructure by backing up too much data.
As in every other aspect of the enterprise, technology is a key asset in the disaster recovery regime, but it's not the only one. Human resources play a vital role, provided they are well-versed in the policies and procedures in your DR plan. If the unthinkable happens, even the most state-of-the-art system will fail if the only one who knows how it works is the guy who built it.