When you start making decisions about your disaster recovery (DR) plan, not only are you trying to prepare for the consequences of losing your entire site to a major disaster, but you are also evaluating the cost of losing a single critical system. How do you know if it's worth the price of setting up recovery systems locally as well as planning for restoration of data off-site? You might come to the conclusion that it's crucial to your business, but the budget might not be there to cover the expenses involved in this level of DR planning.
First, you should determine which systems, if any, would require immediate restoration of services in the event that they were lost while the rest of your data systems remained operational in your original location. Does your corporate email system have to remain up and running no matter what? Is there a billing system in use that, if lost, would render everything else your organisation does useless? These types of systems are good candidates for both local and off-site data recovery methodologies. Often, these determinations must be made by management, which means that you'll need to help translate technical information into non-technical terms whenever possible to ease their decision-making process.
If you do discover systems that need immediate restoration, you'll need to get executive buy-in to begin the process of setting up the proper levels of protection for that particular system. This could mean something as simple as additional disk space for online backups, or it could be as complex as purchasing and setting up redundant hardware.
The determination for how much money you're willing to spend on DR should be drawn in direct proportion to the importance of the system. Get the hard numbers on how long the end users could be without a particular system before a fiscal impact is felt. For example, in many companies, email can be down for several hours — even a full business day — but the company will continue to function. Yes, they will be less effective, but they will survive the outage. Some organisations rely on email so heavily that even an outage of a few hours could be devastating to the business. In those cases, depending only on restoration from off-site tape backups simply will not be an option.
Once you've got the appropriate levels of discovery and eventual authorisation to begin the project, you can start going after the correct budget to obtain the required hardware and/or software solutions needed. It is very possible that you will not get the budget you would prefer, as local recovery of a single system may still be seen as redundancy and nothing more than an expensive insurance policy that you hope never to use.
These are, of course, the same arguments used for denying budget for the entire DR plan in many organisations, so you can use the same techniques to overcome the objections in many cases. Note how much revenue will be lost in the event of a disaster, how long it would take to recover to another location, and how much it would cost to rush-order equipment needed in the event you need to recover locally after the fact.
Local failover isn't always a requirement, especially if you're already planning how to provide off-site restoration. However, in those cases where it is a necessity, failure to plan properly can lead to political and logistical nightmares well before you even begin to figure out what technologies to use.
