Relevant Contents
Need Tailored Business Continuity Insights?
Contact Us Now for Personalized Guidance!
There are many good reasons for an organization to relocate its data center; however, there is only one good way to go about executing such a move and that is carefully.
Successful data center migration does not require that any of the people involved be a genius. It does require patience, effort, meticulous planning, and the ability to tease out the complex dependencies among your applications.
MHA has assisted many organizations with relocating their data centers. In today’s post, we’ll provide some tips and considerations that might be helpful to anyone whose organization is contemplating a data center relocation or is in the planning stages of one.
By way of setting the stage for the rest of the post, here are some of the reasons different organizations find it worthwhile to relocate their data centers: they outgrew their current space; they’re ridding themselves of the burden of providing for the power, cooling, and other needs of a mission-critical computing environment; they want their servers to be in a hardened, more secure facility; they merged with or were bought by another company; they’re taking advantage of the cloud’s economies of scale.
Relocating a data center is a significant challenge but in many cases it is an appropriate choice for the business, whether for economic reasons or to enhance the company’s security and resiliency.
The Key to a Successful Data Center Relocation is Planning, Planning, and More Planning
Data center relocation has a lot in common with controlled disaster recovery. Essentially, you’re gracefully bringing your systems down in one location and then gracefully bringing them up in another, whether it’s physical, in the cloud, or a combination.
Moving the hardware is the straightforward part of a data center migration. What’s hard, and complicated, is understanding the integration of and relationships among the hardware, software, applications, business processes, and impacts if certain portions are not available.
Data Center Relocations Come in Several Different Flavors
These include: traditional (physical to physical with both sites being owned by the company or moving to a vendor managed location), cloud, and hybrid (with part of the data going to the cloud and part to newly relocated company servers).
Partial and full relocations to the cloud are becoming increasingly common. (Note to anyone considering a hybrid data center relocation: think carefully about which processes you will move to the cloud and which you will continue running on company servers. This is a complex topic and one we might discuss in a blog article at a later date.)
Successful Data Center Relocations can be Either “Big Bang” or “Phased”
A big bang move happens all at once in a short period of time. You shut everything down and then the mover comes, you ship your equipment to the new location, unpack it, set it up, and turn everything back on. Such a move might take place over a weekend or a week or two. This type of move is typically only feasible for smaller organizations, where there are only three or four racks of servers. It’s also typically the most disruptive type of move. A phased move takes place in stages and can occur over the span months or a year or more.
What can go Wrong During a Data Center Relocation?
Equipment can break in transit. Your machines might not start up at the new location. You might not have all the dependent hardware and equipment. Processing and application dependencies can be broken, causing business interruptions. Your networking and authentication processes might not be set up. Your allotted time might run out with things still not ready to go. Every single one of these can happen in a move. In fact, they can all happen in the same move—hence determining the interdependencies.
Senior Management Support is Critical
The more senior management supports the data center relocation, the more likely it is to go smoothly. Your company’s senior leadership can assist the move in many ways. One of the main ways is refereeing among the competing desires of different departments. This is commonly an issue in the timing of a relocation. Selecting a time for the move tends to be hard because it almost always happens that a time of year that works well for some departments is a serious inconvenience for others. If senior management is supportive and involved, they can be a great help in resolving impasses between business units.
Planning, Planning and More Planning
The planning phase for your data center relocation is the most important stage of the project. This is where you should spend 90 percent of your time and effort.
Data center relocation is not a job for cowboys, in any of its phases. This project is for careful, patient, detail-oriented people. The lead time for a data center move is typically several months. For a big data center, it might be from six months to a year.
As mentioned above, the key to a successful data center relocation is planning. In the rest of the post we’ll talk about what this planning should look at as well as another key topic, the use of third-party experts who can help your relocation proceed smoothly.
Roles & Responsibilities
The most critical role in the data center relocation is that of technical project manager. This should be a highly skilled individual who is familiar with the technology involved and who has experience in data center moves. If you don’t have such a person on staff, you’ll need to obtain one as an outside consultant.
Remember, the data center relocation is similar to a major construction project in its complexity. The project manager will function like a general contractor, supervising the work of many separate teams and (most likely) outside experts.
You might also have a number of sub-project managers; for example, someone with expertise in moving data to the cloud, if that’s part of your relocation.
Understand Business Needs and Environment
Some of the first things to consider are your project and business needs. Sometimes companies that are relocating their data centers take advantage of the opportunity to upgrade to new equipment. The organization brings the system down on the old equipment then brings it up in the new location on new servers. Such equipment upgrades can be partial or entire, and require careful planning and consideration.
Based on project and business needs, you can then determine the best time to conduct the move. Try to identify the time that will cause the least disruption to the business. You might need to call on senior management to resolve disagreements between departments.
- Your first analysis should aim at coming to an initial understanding of the environment.
- As the project takes shape, start matching your needs against your internal capabilities. Begin identifying areas where you should probably bring in outside assistance.
- As things come into focus, you can start working out how you’re going to do the move and what can be moved when.
- It is critical to understand the dependencies among the various departments and processes, including in scheduling the move.
- Planning for the move should start at a high level then grow more detailed, culminating in a checklist similar to that a pilot uses before takeoff.
If you’re doing a phased approach, make sure you consider the networking, security, and authentication impacts of the relocation. Connectivity needs to be ensured, authentication needs to function, and the applications need to be able to talk to each other. Outside experts can and probably should help you with these assessments.
As you firm up the overall timing, move toward a more detailed level of planning.
- Analyze when the networking needs to be in place, what servers need to be moved when, and all of the technical dependencies that need to be built out.
- When the move starts, you should follow the checklist exactly. Decisions made on the fly with the goal of saving 15 minutes can have costly unforeseen consequences. Only depart from the checklist for a compelling reason.
Seek Third-Party Assistance
In terms of outside help, the role of project manager is commonly filled by an outside person due to its critical nature and the specialized skill and experience required.
Most likely you should also bring in an outside firm to move any hardware that needs to be transported. This should be a company that specializes in the moving of computer equipment, given the delicacy of such equipment and the importance of the task.
You should also get your hardware vendors involved in the move. Notify them well ahead of time. They might have resources to help with issues that arise, whether with hardware or software. You should involve your vendors whether the equipment you use from them is new or existing. They might have special procedures for you to use when shutting things down, or there might be certain components they want you to remove, such as a flash drive or hard drive. They might want to seed the equipment differently in the new location.
Moving a data center is a daunting task but one that can potentially bring significant benefits to your organization. With the proper planning, patience, effort, and resources, such a move can be executed smoothly and efficiently.
Still need help with the logistics of your data center relocation? MHA Consulting provides specific technical expertise to prepare all IT and facilities equipment for relocation.
Richard Long
Richard Long is one of MHA’s practice team leaders for Technology and Disaster Recovery related engagements. He has been responsible for the successful execution of MHA business continuity and disaster recovery engagements in industries such as Energy & Utilities, Government Services, Healthcare, Insurance, Risk Management, Travel & Entertainment, Consumer Products, and Education. Prior to joining MHA, Richard held Senior IT Director positions at PetSmart (NASDAQ: PETM) and Avnet, Inc. (NYSE: AVT) and has been a senior leader across all disciplines of IT. He has successfully led international and domestic disaster recovery, technology assessment, crisis management and risk mitigation engagements.