Everyone recognizes the potential disruption by natural disasters. However, if someone happened to wreck their car into the transformer powering your building, would that be any less disastrous for your company? I’m not sure if you’ve read Cormac Foster’s blog about virtualization and disaster recovery from October – The Impact of Virtualization on Disaster Recovery Planning – but if you’re interested in either topic, it’s worth a read. In it, he talks a bit about the impact that virtualization can have on business continuity, disaster recovery, and some of the planning involved in it.
In his conversation about human resources and planning, Cormac mentions you’ll likely be facing an increase in cost here before you see your ROI. An important cost I’d note for planning on here is services and training. Unless you’ve already got a staff experienced with virtualization or that have gone through the planning process and implementation before, odds are there is going to be a lot that is new. Virtualization by itself changes almost every aspect of your datacenter, from deployment to day-to-day operations to lifecycle management. There are a lot of common pitfalls (and some uncommon) that anyone relatively new to the process will stumble across. Getting assistance in the way of professional services can be invaluable in the terms of scoping your initial implementation, planning phased roll-outs, and even assistance with the implementation itself. Then, getting training to beef up your internal staff’s knowledge of how to implement and administrate the virtualization backbone will pay for itself in the time saved by having to learn on the fly. Having proper prior planning and confident and capable staff will save you money in the long run, and a lot of headaches in the short.
As a follow-on to that, I’d recommend against one-size-fits-all, pre-canned engagements around virtualization and DR. Dell Services follows a customer-centric approach based on standards set forth by the Business Continuity Institute (BCI) and the Disaster Recovery Institute International (DRII) for a reason: one size does not fit all. For example, I get the chance occasionally to speak with customer’s at Dell’s Enterprise Business Center (EBC). A recent conversation with some Caribbean based customers around DR came across staging for their backups, due to the high price of bandwidth from the islands to any mainland DR site. The best solution for them? Essentially, pigeons. At least the same idea – Tier 1 and business critical applications were worth the bandwidth to replicate to the mainland, but most everything else could be backed up to disk and shipped there more cheaply (though on slightly more reliable transport). There are a lot of similarly unique aspects to almost any customer situation – location, local resources, building constraints, for example – that can have a similar impact, and are easily missed in a one-size solution.
As far as budget fights for DR, keep in mind DR is only a part of business continuity – it isn’t ITs to own alone. The single most risk laden decision our Services teams see are businesses making decisions concerning disaster recovery using cost alone as the decision making criteria to the exclusion of the business needs. With virtualization comes a lot of new business agility features that help prove out the ROI of the investment – things like live migration, resource assurances, and fault tolerance capabilities. Our teams view IT Disaster Recovery as a key component in an overall Business continuance Management Program that consists of crisis management, business continuance, and ITDR planning as well as risk assessment/management and linkages to enterprise incident management response at all levels. Properly executed business continuance plans provide the business requirements necessary to make effective decisions on the IT Disaster Recovery Strategy and implementation. In other words, the business itself has defined the worth of the plan, so they are bought-in to the total solution and don’t just view it as an ‘IT issue.’ The request for resources then quickly becomes a much more defensible position when needed, and can be architected properly from the get-go.
Virtualization can provides real disaster recovery benefits, and even more around business continuity, and there is a lot to examine in the planning of it. Care to weigh in, or correct me? What has been your experience? I would love to hear your thoughts on this topic – please keep the conversation going!