Learning From MistakesBy David Strom | Posted 2009-11-02 Email Print
How to keep your company alive in the face of disaster.
Learning From Mistakes
Kelley Drye is an international law firm with offices in a Manhattan building that was damaged after a major steam pipe exploded in 2007. “We couldn’t get in here for six days and didn’t have a good DR plan then,” says IT Director Tom Nohs.
Since then, the firm has implemented a variety of business continuity services, including offsite colocation and cloud-based e-mail. “We were in the planning stages to do a major colocation and cloud-based DR e-mail to a Jersey City, N.J., data center, and we eventually moved all our servers there,” Nohs says. “We also employ Dell’s e-mail management services for doing DR for all our e-mail firmwide.”
Kelley Drye replicates its Microsoft Exchange data across the cloud to Dell’s service. In case of an outage, users can access their e-mail messages by using a Web browser and a secure URL. “I don’t have to change my Domain Name System servers, and my users can be working immediately, giving me plenty of time to restore the services behind the scenes,” Nohs explains.
Any move to a cloud-based DR plan should focus on the support record of the provider. “One reason we went with
Dell’s service is because we have had a long history of using their hardware, and they have a great support model,” Nohs says.
No matter which strategy you pursue, make sure that you test a complete failover to the backup data center regularly. Part of this test is to run with the same loads on which your production systems operate because, otherwise, you won’t catch everything.
New Orleans-based Tidewater Marine tests its DR systems annually, but when the company put its DR site under load, it found that the architecture was insufficient to handle the load, explains John Chaffe, director of IT. “We did not originally build the DR SAN to the exact specifications of the live SAN,” he acknowledges.
Another lesson learned.