Risk Management

A recent survey by the UK’s Chartered Management Institute showed that businesses of all sizes considered loss of IT to be the threat most likely to have an impact on costs and revenue, whilst also showing that it was the most commonly experienced disruption. Given this, why do so many organisations fail to prepare properly for IT disasters?

Well, the fact is that IT disaster recovery is complicated. It’s also about as exciting as watching paint dry, and it ranks very highly on the ‘list of things I’ll think about when I’ve got absolutely nothing else to’. In most organisations, the IT department has NEVER had nothing else to do, so it rarely gets done.

Doing it well requires a lot of planning, and you’ll have to keep on updating the plan if it isn’t going to be totally useless by the time you want it. Testing the plan properly is almost never done, as it’s got a tendency to be ruinously time-consuming, and even has a habit of confirming everybody’s worst fears – that the plan doesn’t do what was intended, and there’s lots more to do. This has left many with either cheap inadequate provisions or worse expensive provisions that fall short of protecting the business.

It’s not all bad, though, because, Virtualisation – the IT buzzword du jour – just might revolutionise the provision of Disaster Recovery. It’s got obvious benefits if you’re starting from scratch (virtual servers are easier to copy around the place for starters), but unless you’re planning on rebuilding all your IT to allow for this, that’s a benefit that you’ll only get slowly, as you replace systems.

More interestingly, it also opens up a new set of strategies for companies big and small who aren’t proposing to completely re-engineer their existing systems. Managed Disaster Recovery services based on Virtualised recovery platforms have the potential to do something quite disruptive to the economics of outsourced recovery.

There are a few reasons for this. The first is rather obvious – Virtual Environments are pretty hardware-independent, so the DR company doesn’t need to keep a load of different pieces of hardware around. They can run large, homogeneous silos, and move workloads onto them at will. Secondly, because the workloads are virtual, the resource itself can be shared between many customers, reducing the amount of money that each customer needs to contribute to run the system.

Finally, there’s another point – virtual environments can be highly automated, and this means that you can do things that just weren’t practical to undertake before – it’s now possible to obtain a service where data is regularly taken from the source servers, applied to a virtual image of that server in another location, and tested on an ongoing basis. And because the images are virtual, it’s not costing an arm and a leg to retain the hardware to do it on.

The real bonus of all of it, though, is the testing – rather than a half-hearted, periodic restore of data from tape to convince the IT department that their backup regime is still working, a managed DR service can know at all times that data is being received, and that the sum total of all of it still makes a functional system.

So that’s nice – maybe your IT department could get on with something useful instead. Like running IT systems, for example.