Careful Planning Leads to Successful Data Center Relocations
- Zhivago Team
- Jun 14, 2023
- 5 min read
Updated: Jun 19, 2023
Data center relocation is basically a “lift and shift” operation, where data center equipment is moved from one data center to another. Data center migration can fall into two categories: moving and/or expanding data and applications to new equipment within a data center (and decommissioning or repurposing the old equipment), or migrating virtual data to the cloud.
The focus of this article is on relocation and the type of migration where there is physical equipment that needs to be moved to or installed in a new location.
Both relocations and migrations are complex. Because of that, data center relocations and migrations can be costly, frustrating, or intimidating—especially for operators who aren’t experienced overseeing significant moves. These downsides can sometimes outweigh the appealing prospect of relocating.
The good news is that many of the issues that may arise during relocations and migrations are preventable—or can be minimized—with proper planning. The difficulty comes in knowing what to plan for and how deep planning should be.
These are the keys to planning a data center relocation.
Evaluate the starting point and the desired result.
Current conditions identify clear areas that could be improved with a data center relocation. Without this complete evaluation of the data center’s current environment and operations, expected improvements from a move would be based on educated guesses, at best.
Conduct a comprehensive review of each piece of equipment, its role within the facility, and determine how it could be improved. If greater efficiency, performance, and cost savings are achievable by simply reconfiguring equipment, the frustration and expense of a move are avoidable.
Document everything.
Every step of the migration or relocation process should be documented, including cable layouts, equipment tagging schemes, and server shutdown protocols. That’s because technicians may have different ways of doing the same task, and memories may fail during a complex relocation. Rather than relying on assumptions or guesses, workers should have a detailed record to refer to for every task they work on. Identify the person on your team that is naturally talented when it comes to detailed work, taking notes, and follow up. They aren’t always the person you’d think of, but they might be your best option to task with the responsibility of documentation.
Anticipate setbacks.
It’s probably not realistic—or even possible—to plan for every problem that may occur during a data center relocation or migration. However, it is possible to learn from the experience of others in similar situations. The extensive history of data center relocations shows common problems that similar facilities have encountered.
These problems are common to almost every data center migration or relocation:
A relocation will put excessive pressure and stress on employees. This can include all the details involved in planning and executing the move, setting up equipment, and running tests, all while balancing these additional responsibilities with regular job duties.
Some details may get left out of the planning process, leading to unforeseen problems.
Data center employees may be skilled in day-to-day operations but lack the specialized knowledge to complete a relocation.
Servers and other data center equipment are sensitive to heat, jostling, vibrations, and impact. Moving several servers from one location to another, regardless of the mode of transportation, may subject this sensitive equipment to at least some of these forces. Some damage is likely and could impact performance and costs.
Legacy hardware, outdated software, or already-damaged equipment may be discovered, making moves more difficult, expensive, and time-consuming than expected.
Relocations are enormous projects. Many employee hours will be spent on making the move rather than on daily operations and maintenance. This can hinder the data center’s overall productivity and performance until the move is completed.
It’s probable that some down time will occur as servers are taken offline and transported. Extensive backups and multiple layers of redundant systems can mitigate this down time, but these systems require their own meticulous plans to ensure success.
Unexpected downtime of one component may create a cascade effect that affects the entire network of components it is connected with. One failure can lead to several.
Incomplete or incorrect documentation causes confusion when putting components back together at the new location. If cable patterns can’t be replicated or hardware doesn’t seem to fit within racks in quite the same way, workers may have more difficulty than anticipated trying to get operations back online quickly.
Remember why the move is important in the first place.
Prioritization is key to devising and executing a data center relocation plan. This is especially true if issues happen when they’re least expected.
If the big picture vanishes, it’s even possible to complete a data center migration and then realize the original objectives were not fully accomplished.
Make a list of the reasons driving the move and document them as a reminder for all employees. This list will help when looking for a new location, but it’s also useful during the takedown and setup of equipment and the arrangement of racks, cables, cooling systems, and other critical infrastructure.
Common reasons to move data center locations often include:
Increased capacity
Lower costs
Better geographical location
Increasing energy efficiency
However, each of these goals are very broad and don’t necessarily cover implementation details. It’s important to drill down into each broad goal to come up with specific items that measure the success of a migration.
For instance, if costs are a driving force, which costs should be cut with the move? How much lower should those costs be? How will the move help accomplish that?
Including the answers to these types of questions during the planning and execution stages can help keep difficult migrations from drifting off track.
Plan for every contingency.
Successful moves occur because of careful planning for as many situations as possible. There should be a detailed outline for each step in the process:
Projecting future growth and needs
Getting all stakeholders on the same page
Finding a new facility
Setting up security measures in the new location
Taking inventory
Backing up data
Migrating data and operations, if applicable
Organizing cables for future use
Hiring professional data center relocation specialists
Safely lifting servers and other heavy equipment
Transporting equipment carefully
Tagging and tracking each piece of equipment
Maintaining compliance to security standards
Arranging the equipment in the new facility
Configuring cables, racks, and cooling systems
Arranging electronic components within racks
Getting everything back online
Running thorough tests to ensure proper operation
Dealing with physical damage, software errors, or other setbacks as they arise
Break the plan down into its individual components and do one thing at a time. This relocation checklist can serve as a guide so nothing is forgotten. View a sample relocation framework to get an idea of what the final plan should encompass.
Meticulous planning will help when something inevitably becomes confusing or frustrating in the middle of the data center move. If a worker doesn’t know where a particular component goes, for example, or a server has errors and can’t access backed-up data, having a detailed, step-by-step framework to fall back on will make all the difference.
Remember, there’s never going to be a great time to execute a project like this. No matter when you do it, it will be disruptive and there will be some pain involved. But once you’ve determined that it must be done, intelligent, comprehensive planning will help ensure that pain is minimized and you can resume operations as usual as quickly and easily as possible.
