I find it difficult to discuss anything worthy of your attention in light of this week's tragic events in the United States. The one subject that surfaces is catastrophic-disaster recovery for Microsoft Exchange servers. After disasters of this magnitude, you can't think about traditional recovery means for your Exchange servers. While rare, these disasters remind us about the importance of recovery measures beyond traditional backup and restore.
How would you respond to the loss of all or part of your Exchange infrastructure? How would your staff respond to such an emergency? The ability to respond to catastrophic events demands that your organization's operations and administration sectors have adequate procedures and training to perform recovery and restore operations. You need to think through an operational scenario for your Exchange infrastructure that will let you recover from catastrophic events.
The capability to adequately respond to debilitating events also requires that you have alternative equipment (e.g., servers, storage, network equipment, power) at an alternative site where you can restore Exchange servers for operation. You also need offsite storage capabilities for your backup media. I've talked to many organizations that surprisingly don't perform offsite rotation of their Exchange backup data. This precaution is crucial if you want to provide disaster recovery for your data and services.
Depending on how crucial your messaging system is to your overall business operations, you might want to consider deploying data replication and mirroring solutions from vendors such as EMC or Marathon Technologies. These products let you mirror data from your Exchange servers to other locations.
These comprehensive disaster-recovery solutions aren't without caveats and they aren't cheap, but if your business warrants such measures, they might be necessary. Obviously, no solution can compensate for the terrible tragedy of events such as those we've seen this week in New York, Washington D.C., and Pennsylvania. However, you need to evaluate disaster-recovery measures that let your Exchange infrastructure operate after such incidents.