News Stay informed about the latest enterprise technology news and product updates.

In disaster preparedness, users say practice makes perfect

Hurricane Ike may have hit the coast of Texas a month ago, but the lessons learned from the storm are fresh in the minds of the IT organizations it affected. In this tip, Deni Connor offers best practices for disaster recovery planning.

Hurricane Ike may have hit the coast of Texas more than a month ago, but the lessons learned from the devastating storm are fresh in the minds of those IT organizations affected.

Consider the stories of these two users:

Roland Etcheverry, vice president and CEO of Champion Technologies, a specialty chemicals company for the oil and gas industry, in Houston, prepares for disasters as part of his daily routine. Champion's offices were damaged by the onslaught of Category 3 Ike, which caused $11.3 billion in damages and left two million people without electricity. Water streamed down walls and windows were blown out but, because of careful and detailed planning, Champion's business operations were unaffected.

In the days before the hurricane, knowing that Ike was likely to hit Houston, Champion IT personnel failed over their business-critical Microsoft Exchange and SAP applications to Scottsdale, Ariz. They recovered those applications by failing back to Houston three days later.

"When the Hurricane Gustav first came through, we started our disaster recovery countdown," says Etcheverry. "Just before it was time to start executing the key elements of it, Gustav turned out to not be coming in our direction. Two weeks later, Ike came rolling through and we started our one-week countdown again."

"Thursday afternoon, two days before Ike was projected to hit Texas, we shut down the data center here and brought it up at our site in Scottsdale," says Etcheverry. "It took us about four hours -- our goal is two -- we are still working on sequencing things to make the failover more efficient."

It was through constant testing of Etcheverry's disaster recovery (DR) plan that Champion was able to sustain no damage to its data center or disruption of its business processes. Etcheverry and his IT team had developed a 300-plus step Microsoft Project schedule to direct their DR efforts.

"We had Microsoft Project running and the entire project was defined minute by minute with all the dependencies," says Etcheverry. "We had a dispatcher who told everyone what to do and a clerk who was recording what was done and we just watched the plan cranking things out. It was uneventful."

A few years before, Champion had installed FalconStor Software's Network Storage Servers, which attach to a storage area network (SAN) and replicate data every hour to the offsite DR location in Scottsdale.

The "real-life" failover of data that Etcheverry executed occurred after numerous controlled failovers in the months before. One failover occurred when the property managers at Champion's headquarters location notified Etcheverry that the power would be cut off to the building during a week-end period.

That failover and failback went smoothly, too, thanks to preparation and planning on Etcheverry's part. Fine-tuning the project and eking efficiency out of his DR plan is part of Etcheverry's daily business. By the end of the first quarter of next year, he plans to fine-tune the plan further by moving to real-time replication.

Tom Comella, vice president of technology and CIO for the Neighborhood Centers in Houston, also sustained power outages in his building for four days, which cut off operations to his servers. Like everyone else in Texas, Comella knew that Hurricane Ike was going to hit on Saturday, so he powered down his servers on Friday.

"We didn't worry about failing over our systems because we use IBM's Remote Data Protection Express software to vault our data to North Carolina," says Comella.

Comella had started planning for the effects of manmade disasters in March 2007.

'We started to think about manmade disasters, like our bank vaults not being available for a long period of time," says Comella. "We just decided that the data needed to get out of the area and alleviate all the legacy problems with tape."

"We thought an offsite backup would be better to get it out of the area," says Comella. And, his data came out of Hurricane Ike unscathed.

About this author: Deni Connor is principal analyst for Storage Strategies NOW, an IT research firm in Austin, Texas. You can reach her at

Do you have comments on this article? Let us know. Please let others know how useful this article was via the rating scale below.

Do you know a helpful storage tip, timesaver or workaround? Email the editors to talk about writing for

Dig Deeper on Disaster recovery planning - management

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.