* Business continuity planning and disaster recovery
Next-generation data centers require next-generation business-continuity-planning and disaster-recovery strategies. Sounds obvious, yet those strategies are an ongoing struggle for the IT executives with whom we work.
Let’s start by defining what’s meant by BCP and DR, and how they differ. In a nutshell, disaster recovery defines the set of tools and processes for recovering from a catastrophic outage – something that disrupts normal business operations. Business continuity planning is the process of ensuring ongoing business operations in the face of disruptive events. Think of it as a continuum, with normal business operations at the far left and catastrophic outages at the far right. Business continuity planning covers everything from the far left to the point at which the disaster or catastrophe unavoidably disrupts operations, at which point DR takes over.
One significant change over the past several years is that point has moved increasingly rightwards. In other words, businesses have become decreasingly tolerant of operational disruptions – at the same time that infrastructure has grown ever more complex. That creates a significant challenge for data center managers, who need to figure out how to provide uninterrupted business operations despite technology and budget limitations.
The best way to do this is to present business managers with a business decision. In other words, data center managers should craft several different “tiers” of service with varying price tags and availability guarantees. That allows them to present business managers with a simple choice: How much business continuity do the business managers want to pay for?
The important point here is that deciding which applications, processes, and sites are mission-critical isn’t an IT function. IT can help business owners prioritize their sites by making it clear what it will cost to “add another nine” to availability and uptime percentages.
Another critical issue data center managers should tackle is determining precisely which business processes are, in fact, critical. One of the most common mistakes that IT execs make is to assume they know the answer to this. Wrong. It’s best to sit down with the business managers and get their perspectives (ideally before crafting the service tiers referred to above). You may discover that an application you thought was mission-critical was something the line of business could live without for 24 hours or more.
Finally, don’t neglect communications redundancy, particularly as part of a disaster-recovery plan. In last week’s newsletter, we spoke a lot about designing resilient data center networks. A critical (though often overlooked) process that needs to be highly redundant is the communications network. Who needs to be notified in the event of an outage? How will these individuals be notified and directed to alternate sites? Fortunately there are a variety of options, ranging from cell phones to instant-message alerts, and savvy data center managers can easily put together a comprehensive and reliable communications network as part as a bulletproof BCP/DR plan.
