The more critical a project is to a business, the greater attention it deserves.

Which is why it puzzles me that many companies have little or no appetite for discussing the most important IT project of them all: disaster recovery. DR is the set of policies and procedures that will allow vital technology and infrastructure systems to recover (or continue operating) in the event of a disaster.

A disaster could happen through physical events such as fire, flood and hardware failure. But it could also occur due to a software upgrade gone wrong, or human error.

The greater the risk, the more businesses should do to protect themselves.

Note to self: Don’t mention the mksysb

As an AIX guy, when I first come into contact with a system, I very quickly check the basics, such as when the last mksysb backup was taken. However, in a large organization, when you’re speaking with senior management, they may not know that you back up the AIX operating system with the mksysb command.

So instead of asking something like “when was your last mksysb?”, I find it more effective to centre the conversation around how critical the IBM Power System is to the business.

So I ask questions like:

“Is your ERP dependent on that IBM system? What would happen if you lost connectivity? Who would you call? What would happen next?”

And then the conversation quickly moves into the cost of extended downtime. If a company’s production system is down for a week – perhaps due to site failure from a flood or a fire, for example – the business impact is enormous.

Asking Business Questions

The conversation naturally leads into questions about how much the business would suffer, so we’ll look into questions such as:
“would you still be able to trade?”
“When would you consider laying off staff?”
“What about your distribution channels?”

From my experience, the discussion quickly moves towards relationships with other businesses (suppliers and customers). If they’re a high-profile company, then brand damage is going to be high in the minds of senior managers. Think of the customers whose orders don’t arrive, and who resolve never to buy from that company again.

Business impact from unexpected downtime is very, very significant. For some companies, a day or two of downtime would be painful, but for others, they would simply be out of business.

As a way of leading this conversation (without mentioning the AIX mksysb command!), I’ve prepared a risk self-assessment checklist. It’s a way for management to have the conversation internally about just how prepared the company would be if their IBM Power System environment became unavailable.

It makes a lot of business sense to protect your mission-critical systems before discovering that your recovery plan was not ever going to meet business requirements.