Relocating a data center for business continuity or disaster recovery purposes is never an easy task. If you are planning to relocate your data center, many situations could thwart a smooth on-time move. You'll have critical data assets scattered throughout the environment. In addition, business mergers, acquisitions, spin-offs and major market changes happen all the time, and these could also affect a move. If you thought that disaster recovery plans for data centers are only applicable when the center is operating normally, think again.
In this tutorial, learn data center relocation tips and strategies, and how to avoid common data center relocation problems.
A TUTORIAL ON RELOCATING YOUR DATA CENTER:
TABLE OF CONTENTS
Data center relocation strategies: Key steps
Step 1: Full IT survey and assessment
Step 2: Detailed analysis and design
Step 3: Data center migration plan
Step 4: Risk identification and mitigation
Step 5: Implementing the data center relocation plan
The role of business continuity/disaster recovery in the relocation
One of the key aspects of data center relocations is being able to not only anticipate potential disruptions to the move, but also to have backup plans ready to activate if a problem does occur. If it needs to be modified to address potential relocation issues, be sure to update it before the move occurs.
Data center relocation strategies: Key steps
What are the key steps to take for successful data center relocations?
- Conduct a full survey and assessment of the IT environment from the outset
- Complete a detailed analysis and design for the move
- Prepare and approve a complete data center migration plan
- Identify key risks and initiate an appropriate risk management plan
- Plan and execute an overall implementation strategy
Data center relocations can fail for many reasons. As moves do not usually happen overnight, intervening events often arise that can disrupt relocation. For example, people come and go, key staff may leave, and businesses may merge or acquire other companies. Insufficient space or issues with electric power may force an urgent restructuring or sudden change of plans. Senior management may suddenly slash budgets cut costs, possibly in the aftermath of disappointing financial returns.
One thing is certain: data center relocations always pose a serious threat of disruption to operations or data loss that can impact crucial business functions and relationships. IT managers must understand this fact from the start and recognize that such disruptions and data loss are unacceptable risks. While the opportunity presents itself, a disaster recovery plan review prior to a data center relocation is an important step towards ensuring a successful move. Let's examine the steps in more detail.
Step 1: Full IT survey and assessment
Begin data center relocation preparations with a comprehensive review and survey of the current data center. Previous data center audits can be very helpful. The research should include an inventory of all software and hardware, servers, and storage and networking equipment. It also should identify all stored data and classify it according to business criticality. In addition, be mindful of the operational requirements of the current data center, e.g., space, utility services, cabling, lighting, raised flooring, security and ensure they are factored into the move.
For example, if your data backup policy is to send daily backups to an offsite storage facility, you may need to make special arrangements with the service to address changes in the backup process during the move, and to ensure that critical systems and data are as current as possible.
The assessment should also identify and document the organization's technical and business requirements. You want to determine the key services the data center provides, establish business priorities, and identify any planning and budgeting constraints for the data center relocation, such as saving money by moving existing furniture as opposed to installing new furniture at the new site. This could present potential scheduling and operational issues if the furniture is not delivered on time. As such, you'll want to make sure the moving company has backup options available in case there is a vehicle breakdown or accident.
Every data center move presents risks, and any business interruption directly or indirectly caused by the move could affect your bottom line. So in addition to conducting assessments of operational requirements, such as equipment and furniture, it's also a good idea to conduct an analysis of the impact of IT disruptions on the business. This analysis can help prioritize business services for re-launching at the new facility, and establishing acceptable downtime parameters in case of a disruption.
During this stage, define potential risks and prepare BC/DR and contingency plans to mitigate possible problems during the relocation. If it's feasible, there should be more a detailed migration plan that addresses specific risks, such as accommodating utility company delays and issues with moving companies. The ideal strategy is to contact key organizations, such as utility companies, well in advance of the pending move and include them in the planning process.
Next, the relocation team begins to assemble all previously collected data and documentation into a structured and inclusive data center migration plan. This includes a complete in-depth risk analysis, a key part of the BC/DR process, and it should focus specifically on the move. It also encompasses floor layouts and rack diagrams, specifies the exact tasks required to complete the relocation and assigns responsibility for each task. Finally, it charts the migration priorities in a matrix format. Since data systems most likely contain complex interdependencies between business-critical applications and assets, the relocation team should map these dependencies with as much detail as possible. The plan needs to factor in guidance from utility companies, particularly telecom and power, as well as technology suppliers and outside specialists, such as equipment movers, electricians, and heating, ventilating, and air conditioning (HVAC) consultants.
Throughout this process, the relocation team identifies both technical and business risks. Having completed this activity, the teams should assess, classify, and prioritize the identified risks for the purposes of mitigation. Some risks may be deemed minor or not worth the cost of mitigation. Other risks will require a full mitigation program with business continuity/disaster recovery and contingency planning. In some instances, certain risks may even raise insurance considerations.
In this final step, the entire relocation effort comes together. People perform their assigned tasks, and vendors commence their activities. At this point, the relocation team brings in the technical experts and professional systems movers. The team makes the necessary arrangements with the hardware and software vendors for proper shut down and start-up of various systems. The team also arranges for moving utilities from the current location to the new data center and restarting them. Essentially, the data center should be moved as quickly and efficiently as possible so as to minimize disruptions to business operations and critical processes.
While the process of relocating a data center is far more complex than suggested here, the need for emergency and backup procedures and plans in case of a disruption is no less critical. Among the risks are the following:
- Loss of power
- Loss of data
- Severe weather that impairs moving activities
- Problems with transporting equipment
- Damage to equipment while in transit
- Utility services not ready at new space
- Inability to power up systems and reload applications
- Lack of staff to manage contractors at new site
- Lack of staff to complete the move
- Union grievances associated with moving company, installation firms, cabling firms
The above list is only a starting point for the potential risks to a data center move. It may not be possible or practical to have business continuity/disaster recovery or contingency plans that address all of these and other issues that may arise. Your best alternative, assuming you have no detailed BC/DR and/or contingency plans, is to have a relocation plan with detailed processes and lists of contacts, so that you can quickly improvise alternate solutions. Examine existing business continuity/disaster recovery plans to identify any additional technical, operational or staffing issues that may affect the relocation, such as moving a call or contact center. As the move progresses, schedule regular checkpoint meetings to determine what risks have surfaced, if any, and how they have been mitigated or otherwise handled.
Once the move is finished, conduct a detailed post-event review to see what worked and what did not, including how the organization dealt with incidents that impacted the move. Review and update all relevant data center business continuity/disaster recovery plans as soon as possible after resuming normal operations. Exercise updated BC/DR plans if needed, and keep them up to date.
Moving a data center entails many risks, some of which we have discussed in this article. Advance preparations, including business continuity/disaster recovery planning, can ensure a smooth and uninterrupted move.
About this author: Paul F. Kirvan, FBCI, CBCP, CISSP, has more than 20 years experience in business continuity management as a consultant, author and educator. He is also secretary of the Business Continuity Institute USA Chapter.
This was first published in July 2009