Many attendees at this week's Storage Decisions Disaster Recovery Planning Seminar are prepared to spend as much or more on DR this year as they did last year. However, they identified cost as their chief obstacle in filling all of their disaster recovery spending needs in 2009. Some IT pros also acknowledged that their current disaster recovery planning is lacking.
Steve McDermott, IT operations manager at London-based insurance company HCC International, said his business was increasing disaster recovery spending after recognizing that the existing DR budget was inadequate.
"I've been in the company two years and we didn't have a successful DR test until last week because we've never had sufficient spend in that area," McDermott said. "But the company sees the need for DR capability and so we have an increase in budget this year."
He added, "The key obstacles are that some of the business users see DR as a cost and that [DR scenarios] are never going to happen, so why spend money on it? On the IT side, the obstacle is determining how we replicate our data and finding the key technologies to use."
Complex infrastructure, compliance drive DR spending
In some sectors, where real-time processing is critical, extensive DR strategies have been in place for some time. In banking, for example, the challenge is to stay on top of a constantly changing and complex infrastructure.
"Our budget on DR is increasing, and that's driven by rising storage costs and complexity in applications," said Andrew Reid, IT director at London-based Credit Suisse. "Cost justification and application design are the key obstacles to achieving what we want to achieve."
Heightened compliance was also a key factor in increased disaster recovery spending for some financial services, where data growth can be an obstacle toward achieving DR goals.
"Spend is increasing as the business becomes more aware of what could go wrong. We're finding the business is coming to us to look for a more robust DR solution," said Gary Molloy, London-based operations manager at financial services giant AIG. "Data growth is the main obstacle to achieving what we want in DR. If a disaster happened today, I don't have the capacity to restore everything straight away."
Of course, data growth is a challenge in many industries.
"Spend is staying the same, but we pretty much proved that our DR processes don't work so it's time to re-analyse what we do," said Robert Atkinson, IT manager at London TV production companies Sunset & Vine and Mentorn. "The first test of our DR processes showed there were big gaps. What's happened is that we've grown so quickly that what we put in place in the past is no longer sufficient for our needs."
Atkinson's situation isn't unique.
"Our DR strategy isn't very good at the moment and it' something that we're looking at," said Nick Wright, project manager at publisher Random House. "DR tests haven't gone well and we rely on an external provider to bring trucks in; it's not a very recoverable situation."
He added, "Cost and bandwidth are the main obstacles. We're trying to get the RPOs [recovery point objectives] and RTOs [recovery time objectives] down as much as possible while bearing in mind how much some of the technologies cost."
Challenges to a workable DR strategy
A recent SearchStorage.co.UK purchasing intentions survey, which surveyed 435 data storage managers about their information technology purchasing intentions, revealed that disaster recovery spending is set to increase in 41% of shops surveyed while an additional 32% expect to maintain their 2008 levels.
Jon William Toigo, chief executive officer and managing principal at Toigo Partners International, as well as founder of the Data Management Institute (DMI), was the keynote speaker at the Storage Decisions Disaster Recovery Planning Seminar. He said disaster recovery spending often keeps rising no matter how dire the economy. Toigo added that the key challenges in creating a workable DR strategy are getting senior management buy-in, matching data to the level of protection required and implementing an effective testing regime.
"You have to help management understand what the objective of the continuity capability is," Toigo said. "[It's] not to go out and buy a bunch of gear and deploy it, but to integrate self-healing characteristics into business processes and to obtain information that can be leveraged in a variety of ways to help the company's objectives."
He added, "You have to know what data it is you want to protect. Often, nobody really has their head around the data asset, and that's what you're trying to protect. So you have to classify and define it. The payoff is huge," he said.
"But it's very rarely articulated to the decision makers in the front office, so you don't get management buy-in. And if you don't get that, you don't get disaster recovery," Toigo said.
"You have to plan to test because testing is the long tail cost of disaster recovery planning and you have to bring the cost way down, for example by bringing in the testing of replication on a day-to-day basis."