Mission-critical now a meaningless phrase

Symantec has just released its 2008 Disaster Recovery Research Report, which surveyed 1,000 executives globally on DR (disaster recovery) trends, of whom apparently 5 per cent were Australian.

While I'm wary of placing too much emphasis on statistics that only involve 50 locals, two figures did stick out. Firstly, the number of C-level executives on corporate DR committees has dropped from 50 per cent to 33 per cent.

Perhaps we need some nastier natural disasters or terrorist attacks to refocus everyone's attention.

Secondly, the percentage of applications defined by the surveyed businesses as "mission-critical" is now an astonishing 64 per cent.

I'd venture to suggest that if you think two-thirds of your IT is mission-critical, you're either running an incredibly lean and efficient operation (not likely given the universal complaints about keep-the-lights-on budgets) or you haven't got a clue how many applications you have and which ones you need to manage (all too likely).

A disturbing sub-trend pointed out by Symantec's Australia and New Zealand director of systems engineering Paul Lancaster is that many businesses would define a core application as mission-critical, but not accord the same status to underlying systems (like databases) needed to keep those systems running.

Given that, it's not surprising that when these businesses ran DR test scenarios, a good third of them failed.

Cluelessness is also evident in the rush to use virtualised systems as a cheap-and-cheerful DR option. "Virtualisation has sprawled instead of grown," Lancaster said. "People have seen apps being able to be put on a single server and the management side of it has been more or less left up to the application owners."

That's a real (non-virtual) disaster waiting to happen.