For a Disaster Recovery (DR) strategy to be effective, there must be a second data center environment that is in a different geographic location, sufficiently far enough away from the primary site to minimize the possibility of both data centers failing simultaneously. The premise is that your production environment, or primary data center, must be continuously replicated to a safe and secure second data center. In the case of a critical disruption that threatens the continuity of your business, the primary data center quickly fails over to the second data center, ideally without users experiencing any delay, with minimal service disruption and minimized data loss. The features, capabilities, location, and configuration of your second data center depend on your requirements and the availability of appropriate software solutions, infrastructure, and resources to meet those requirements.
Let’s look at the most common scenarios companies encounter when implementing a second data center in the quest for the perfect DR solution.
View top insights into data security and how the right partner can ensure your organization doesn’t stumble.
The company has a second data center offsite that it built and has been managing on its own. Owning and operating data centers is costly, does not generate revenue and because its operation typically falls outside core business operations, the company is not equipped to operate it sufficiently. In many cases, these companies end up eliminating the funding for the existing, outdated data center and replacing it with a third-party vendor. The vendor provides updated IT infrastructure with newer DR technology, but most likely proprietary, which means the company has limited options for customizing the solution to its needs.
The company is developing a DR plan but does not have an existing second data center. The plan is to build a company-owned data center by repurposing old equipment that was to be decommissioned as a more cost-effective approach. Beware, this strategy is flawed in two significant ways. In trying to minimize purchase costs by leveraging older equipment, future costs will be incurred to update the hardware and software over time as business needs and demands change. Additional IT staff and training may also be required to accommodate these changes. Second, older equipment is decommissioned for a reason, most likely because it can no longer perform as needed. If your DR solution is supposed to be the guardian of your business, you want state-of-the-art technology behind it. That way you will have the ability to deliver your customers and end-users a seamless experience that seems as if you were operating out of your primary, production data center in the event of an actual disaster. This is particularly important when considering large scale disasters that require a prolonged, potentially multi-month repair or rebuilding at the prior primary data center. In this situation, the DR facility becomes the primary facility for an extended period of time.
The company needs a second data center, and for cost reasons, decides to go with a data center colocation provider. On the surface, this scenario may seem to make more economic sense than owning and operating your own second data center. However, the capabilities, options, and certifications of the third-party provider must be carefully reviewed to ensure that all of the customer’s requirements can be met throughout the course of the long-term agreement. It’s also important to keep in mind that the costs (both recurring and non-recurring) from the third-party provider may exceed the overall cost of owning and operating your own secondary facility. Fully understanding the service agreement terms and conditions, as well as the billing model, is essential to accurately forecasting the total cost of the third-party relationship over time. Aside from variable costs, working through a third-party provider may cause delays in fulfillment when a change is required. As business needs change, some providers may not be flexible enough to accommodate the new requirements in a timely fashion.
The strategies described above are myopically focused on cost savings while sacrificing quality technology and expertise needed to address today’s threats. Let’s review the options for setting up a second data center and their advantages, along with their disadvantages.
- Build your own – This option is expensive because it requires the company to have the budget to build and operate a data center facility. This includes purchasing the hardware and software, and investing in dedicated IT staff to both operate the facility and manage the DR responsibilities. This option is usually only feasible for large enterprises. However, the benefit is that your security strategy remains under your direct control.
- Data center provider – The provider supplies the space, equipment, and software. Often, the provider is associated with a specific vendor and its products. Rarely do the best solutions comprise technologies from one manufacturer. Whether operations personnel are on your staff or the provider’s staff, they must be experienced in all deployed technologies no matter the manufacturer.
- Traditional vendor – Much like the data center provider, the vendor provides the facility and most definitely offers only its own software and hardware. For the company who values working with a one-stop-shop, this could seem like an attractive option. However, no single vendor provides the best available technology to handle all facets of DR.
- Hybrid approach – With this strategy, the company secures the space and then works with a DR provider who provides the equipment. This approach offers cost savings by going direct to acquire the necessary real estate, and also gives access to the technology best suited to your unique solution. The subtle disadvantage is that you might not have full access to a best-of-breed solution.
- Public cloud – The company deploys DR technology that replicates to the cloud. The lower operating expenses are compelling, but this strategy requires that your staff has the right experience to monitor, test and manage the solution.
- Managed DR service – A managed service provider architects the solution with the technologies best suited to your business objectives. Because the provider is often bound by SLA agreements, the equipment and processes are linked to performance rather than to a specific vendor which elevates the quality of your DR solution.
Is One Better than the Other?
With many of the options, there are tradeoffs. For example, the most attractive option might be to build your own second data center and maintain control of operations and security. In turn, you would be spending a significant amount of your budget and allocating valuable resources to the company-owned solution. For many companies, the total investment and ongoing costs associated with this approach make it financially unattractive and most likely, infeasible.
Maybe leveraging public clouds for a DR strategy is the answer. After all, the approach has been gaining popularity over the past 24 months. The concept is particularly compelling because of the assumption that this approach will certainly reduce costs. However, this is not always the case and the trade-off is that you are relinquishing much of the control to the cloud provider so you are limited in managing your recovery time objectives (RTO), making it inappropriate for certain needs. For example, you may discover that the storage is not sufficient for the I/O demands of your systems and applications as you spin them up. If this is the case, you will be faced with having to pay for a higher storage tier which dispels the myth of potential savings.
As far as the other options we mentioned? They have similar trade-offs and settling for something less than ideal is not something you want to do with a DR solution. However, there is one option that requires no compromise. A managed service provider can tailor the solution, because your ideal DR strategy depends on your requirements, your budget, and the value you place on your business continuity. This type of partner knows the precise technologies for the job, and has expertise with best practices including comprehensive testing and continuous monitoring to ensure compliance with RTO and RPO objectives, as well as industry regulations.
“Cybercrime caused 22 percent of data center outages in 2015, reported in a recent survey conducted by the Ponemon Institute and sponsored by Emerson Network Power.”[i] Now is the time to up your game. Whether you want to build your own data center, leverage public or private cloud options, or are engaged with a third-party vendor, a managed DR partner with the right expertise and capabilities can work with you to design and deploy a solution that best meets your requirements – with no trade-offs.
For further information, refer to Data Recovery as a Services (DRaaS).