As businesses rely more heavily on their data center services, the cost of downtime increases. And for these (and other) businesses, hurricane season is a strong reminder of the need to plan ahead for potential disasters. One disaster recovery option for companies is a remote facility that can take over operations in the event that the main facility suffers a downtime event. But is such a strategy cost effective?
Evaluate the Cost of Downtime
The answer to the question “Is X worth it?” is almost invariably “It depends.” The case of disaster recovery data center sites is no different. For one company, a remote disaster recovery site might be worth every penny of the expense, and then some; for another company, it might be a cost-prohibitive option that cannot be reconciled with the current (or future) financial situation. The first step in determining if a remote site—or any other part of a disaster recovery plan—is worth the cost is to evaluate the cost of downtime. This is a topic that could be discussed at great length, but the following are a few considerations to examine when determining this cost to your business.
- How much direct income will you lose if your data center fails? This is the most obvious cost: if your data center is part of your link with customers, then downtime means lost revenue.
- How much employee time will be redirected from productive efforts to recovery efforts following a downtime event? Every man-hour spent fixing the problem is a man-hour not spent developing new products, providing services to customers or otherwise building the company.
- How will downtime affect your company’s reputation with existing and potential customers? A poor reputation for reliable service means customers will go elsewhere. Will a disaster that takes down your data center be something your business can recover from? Or is ongoing service continuity a requisite for success?
- How much will emergency services and equipment cost your company? Getting your data center back online after a disaster can be very expensive if you try to do it quickly. Emergency service is expensive, and getting replacement parts or devices in short order can also carry greater cost.
Again, these are just a few considerations. And how each company evaluates its potential losses owing to downtime will vary by industry, clientele, criticality of services provided and so forth.
Once you have determined a working number regarding downtime cost (say, dollars lost per minute or hour of data center downtime), you will be in a better position to evaluate disaster recovery options, such as remote sites.
Benefits of Remote Disaster Recovery Sites
In the event of a major disaster, all your efforts at redundancy, data backup and so on to keep your data center running may be in vain. A hurricane, for instance, might knock out utilities such as power and water for an extended amount of time, and they can even cause significant damage directly to your facility. Even worse contingencies are certainly conceivable. One logical option to circumvent the risks to your business is to build a remote disaster recovery site: another data center that can take over operations if your main site goes down for some reason. Locating the backup site remotely reduces the risk that a disaster affects all of the company’s data center resources simultaneously, enabling continuous service to customers and to the company itself.
A disaster recovery site is another layer of redundancy beyond backup generators, redundant equipment in the main facility and so on. This site can be a company-owned facility—perhaps on another company campus—or a colocation or similar facility that houses IT equipment for a variety of customers. Furthermore, the remote site may operate in a variety of modes, ranging from “cold” sites that must be loaded with software, data and so on when a disaster strikes to “hot” sites, which operate continually as full data center facilities and are thus able to take over operations on the spot in the event of downtime at another facility.
Costs of Remote Disaster Recovery Sites
The costs associated with a remote site vary depending on the type of implementation: obviously, for instance, a full company-owned data center will involve much greater capital cost than a colocation-based implementation. In the former case, all the same costs associated with building a data center apply, ranging from real estate costs to construction, IT equipment and personnel costs. Colocation implementations are less, but they still involve capital costs for the IT equipment as well as ongoing charges, even if you don’t run most or all of your equipment.
Costs will vary depending on the distance of the remote site from your main site. Disaster recovery sites should be far enough away from the main site such that a single disaster has a low probability of affecting both locations (and that probability depends on the kinds of threats present in the particular area). On the other hand, locating a remote site too far from the main site has its own difficulties: moving staff to the backup site can be burdensome or even impossible, but maintaining remote staff can be expensive.
The question is then whether the cost of downtime at your company’s main facility can justify the cost of constructing and maintaining a remote disaster recovery site. For many small or even medium-size businesses, the answer may well be a resounding no. Nevertheless, for many other companies, the potential returns from maintaining a remote site may well outweigh the costs of downtime. Again, the answer to the question of whether a remote site is worth the cost depends on the situation each company faces: there is no universal “yes” or “no” answer to this question. Furthermore, even the “yes” option raises a number of possibilities for the remote site, ranging from how it is operated to what kind of facility it should be.
Alternatives to a Traditional Disaster Recovery Site
Even if your business necessitates a remote disaster recovery option of some sort, you are not locked into buying land, building a brick-and-mortar data center of your own and then staffing it with personnel ready to take over operations at a moment’s notice. A number of alternatives can help you reduce cost and still gain the protection of a backup facility, should disaster strike. One unique possibility raised in Data Center World (“Five Cheap and Easy Ways to Prepare for Disaster”) is to “share” disaster recovery responsibilities with another, similar data center. Two companies could, for instance, each act as the other’s disaster recovery site, saving both the tremendous expense of building a new facility. Another option that forgoes capital costs entirely is a cloud implementation, although this approach also eliminates company control of the physical IT resources. And companies can also mix and match portions of the various options to create a solution that provides the best balance of cost and effectiveness in the event of a disaster.
Arguably for most companies, the range of options for disaster recovery means that a fully company-owned remote site mean is unnecessary—particularly if the cost is prohibitive. Depending on the cost of downtime, however, some companies may find that a remote site is the best option. Whatever the case, however, you must evaluate the cost of downtime to your own company and compare it with the available options in light of your budget. Even though one disaster recovery option may be “better” than another, that doesn’t mean that it’s better for your company.
Photo courtesy of Tom Raftery