The past several years have produced plenty of motivation for enterprises to institute broad disaster recovery programs, so much so that it's hard to imagine even a medium-sized organization without one.
Unfortunately, many plans turn into the IT equivalent of a mutual fund: Once the initial investment is made, it's pretty much ignored until something goes wrong.
As our own Rob Enderle pointed out last week, the swine flu pandemic is a reminder that enterprises can be laid low by threats to the workforce as well as destruction of physical infrastructure. So if your recovery program does not have contingency plans for loss of key personnel, it's probably time to review it.
The fact is, though, that there are any number of ways in which enterprise service can be interrupted: earthquakes and other natural disasters, fire, human error, even civil unrest or revolution. So it's no surprise that many of the leading DR vendors are coming up with ingenious ways to counter as many possible threats as possible with an ever-widening set of tools.
Data Domain, for one, just added a number of new features to its Replicator software system designed to help maintain business continuity on a global scale. The system now supports 90-to-1 remote site fan-in to a single controller, meaning enterprises now have a way to institute automated deduplication around the world. The company also boosted the system's replication and mirroring capabilities with greater end-to-end bandwidth for speedier recovery.
Stratus Technologies, meanwhile, has upgraded its SMB-targeted Avance software with what it calls Split-site capability that allows paired x86 servers to be separated geographically to protect against localized events. The system also sports new Web-enabled remote systems management, iSCSI SAN support for the Dell EqualLogic PS5000E, and support for internal RAID 0, 1 and 5.
DR is in such high demand these days that other data management firms are getting into the act. CaseCentral, which specializes in eDiscovery tools for the legal industry, recently added a business continuity planning (BCP) protocol to its portfolio designed to maintain service in cases where processes fail but IT resources are still intact. The company maintains a secondary data center with 500 TB of storage that is designed to restore client processes, documents and metadata within a 90-minute time frame.
It's important to remember that, despite the headlines, the risk of cataclysmic disaster is still very small. But that doesn't mean there is no cause for concern. Even minor events can cause major problems if contingency plans are either non-existent or are gathering dust in a storeroom somewhere.
The experts will tell you that the smartest move right now is to institute a plan and subject it to routine drills designed to both reveal weaknesses and establish individual responsibilities in the event of a real emergency. It's impossible to plan for every contingency, of course, but at least the preparation could prevent a minor problem from turning into a major disruption.