Keep it covered

DURING AN INTERVIEW, an actress was asked where she was last Sept. 11. She was rehearsing a play in New York, and when the news of the day's events came in, the director sent everyone home. They all returned to the theater the next morning and resumed their regular schedule. That's how most of us handled it. Going back to work was the right thing to do for many reasons.

Free IT resource Hear how top CIOs turn change into a competitive advantage. Sponsored by HP Free IT resource Try Sun servers, workstations and storage products free for 60-days. Sponsored by Sun Microsystems

The unexpected duration and depth of this economic downturn creates a more complicated dilemma for IT: how to fit business continuity into the list of priorities. It's difficult enough to decide which projects to idle and which employees to let go; choosing which of many top priorities to table is an unpleasant, contentious task. It might be left to chance or brushed aside with a vague directive to cut "everything not necessary for basic day-to-day operations."

There are some corners IT can't afford to cut. A few, such as security, are obvious. Business continuity often gets the ax because its return is unpredictable -- you cut back on insurance when the budget is tight -- but it's critical to your company's survival. Without preparation, a failure can bloom into a costly data or service loss.

Continuity has to be a flexible concept in unstable times. Maintaining even a reduced state of readiness requires creativity and a degree of detachment. When allocating manpower to business continuity tasks, recognize that any assignment given to an individual might have to be reassigned later. Pool your staff so that the next item on the task list is handled by the first available person. Put all able people in the pool (hierarchies are a luxury at such times).

If your equipment budget is frozen, be prepared to shuffle your assets. Continuity is a higher priority than performance or capacity; you might need to take some servers out of a farm to act as hot spares, or reallocate networked storage to create mirrors. Continuity processes that are self-activating and unattended, such as server and storage fail-over, are ones you can count on when you're understaffed.

Also, push preventive maintenance and testing up the priority chain; increase the frequency of tape-drive head cleanings, UPS battery testing, and failure drills. That might require some scheduled downtime, something many companies have eliminated, but continuous availability is less important than resiliency.

It's tempting to send a message to management about the importance of continuity plans by allowing things to fall through the cracks. If your boss tells you to fire half of your administrators, shouldn't you teach him a lesson by letting his critical backups go stale? But look at things from the other side: Your boss is probably as frustrated by the lack of resources as you are, and just as helpless to change the situation.

Skip the futile protest. Simply stated, everyone is in an impossible situation. Find ways to get things done with the resources you have and speak plainly about what you can't do.