Fifteen best practices for a successful data center migration

Data center migrations are often complex and risky. The following best practices will help organizational managers invest the right amount of time and money in planning, execution, and testing to ensure normal business operations and maximize the chances of a successful data center migration.

Key challenges facing organizations

Colocation, merger and acquisition activity, obsolete facilities, consolidation initiatives, and new methods of increasing data center capacity often require the migration of IT equipment from one data center to another.

Most data center migrations result in cost overruns and reduced performance due to poor planning or a lack of dedicated teams and resources.

Existing staff often lack experience in planning and organizing the migration of IT equipment from one data center to another.

Planning for Data Center Migration

Organizational leaders plan data center migration as part of their infrastructure delivery strategy:

• If an enterprise does not have the expertise and skills to effectively plan, manage and execute a data center migration project, then collaborating and obtaining external expertise is an effective approach.
• Design and execute a strong communications plan during all phases of the project, engaging frequently with all affected publics and providing complete information.
• Develop risk mitigation, migration, and fallback plans by testing with disaster recovery (DR) procedures.
• Continuously document the entire project process to develop lessons learned and reference material for later evaluation and closure.

This research examines best practices for data center migrations. This is not a comprehensive review of all migration methods. Rather, it is a list of best practices designed to help business leaders succeed during data center migrations.

Data center migration is more than just setting up an infrastructure and moving workloads from point A to point B. It is often a complex and risky project, and having the right processes and expertise is critical. Having a perfect plan is essential to a successful data center migration. However, if the right resources are not in place within the organization to develop the right plan, these internal resources must be combined with external resources to effectively plan, manage, and execute the migration project. Business leaders must prevent data center projects from adversely affecting the business. They must ensure that applications and services continue to operate normally to ensure minimal downtime and no performance degradation.

To help business leaders achieve this goal, the following content focuses on 15 best practices for data center migration.

15 best practices for data center migration

Most successful data center migration projects have similar practices, such as expertise, preparation, management, execution, communication, and business alliance. The purpose of this study is not to conduct an in-depth study of each of these practices, but to provide a pragmatic list of best practices. This comes from years of observation and summary of data center migration project practices.

Based on the steps to a successful data center migration, the 15 best practices for data center migration are organized into five main steps as shown in Figure 1: start, risk assessment, planning, execution, and shutdown. However, organizations may change the order of these best practices based on their migration situation. They can also apply these best practices to multiple environments (for example, local data centers, colocation data centers, or cloud computing).

15 best practices to consider for your data center migration project

Here are 15 best practices you should consider in your data center migration project.

1. Master skills and expertise

It is critical that your staff has the right skills and expertise. Since data center migration is not part of daily work, existing staff often lacks experience in planning and organizing the migration of IT equipment from one data center to another. Business leaders must evaluate whether their internal resources are available and can effectively plan, manage and execute the migration. If internal resources are lacking, it is recommended to work with partners to adopt external expertise and skills.

Note: One of the main reasons data center migration projects fail is that the leadership team often views data center migration as an equipment migration project. However, most of the work and most of the risk of a data center migration lies in developing a plan for migrating the workload. Migrating the equipment is the easy part, while workload placement, dependencies, business impact, and risk are the hard parts.

2. Establish a project team

Major or complex data center migrations require a dedicated manager. This manager should coordinate with all relevant representatives and build a team. Management should have the authority to allocate resources and personnel.

Not only should this team member be an expert in their field, but they should also be willing to learn related disciplines, as it is critical to resolve the interdependencies of all equipment. The more detailed or extensive the plan for the data center migration, the more likely it is that team members will need to devote full-time hours to the work. And these staff members' experience and skills should be practical and applicable, and they should be ready to improve their skills with outside help. For longer projects, be aware that changes to the plan may be needed: during the project, when team members move into other roles, their skills will need to be updated.

3. Prepare your project

A successful data center migration depends on previous preparation and high-level planning. Enterprises should establish a committee that includes IT and business representatives to fully consider all necessary resources, including:

• Migration costs
• Existing contracts for software, hardware, services, maintenance, disaster recovery, facilities, and other projects

Project preparation should also include the following tasks:

• Create a detailed inventory of devices and applications
• Conduct a Business Impact Analysis (BIA) review

These basic tasks should include a detailed assessment or audit of exactly when the migration is needed and how it will be done.

Rehearsing the migration is also an integral part of its preparation. To ensure that all departments understand the process, organizations should first define the drill on paper and iterate as necessary. Then, organizations should schedule multiple migration drills during the project to verify assumptions and determine key information such as migration time and resource availability.

4. Simplify the process

The less things to move, the easier the migration. So simplify, minimize, virtualize, consolidate, and eliminate as much as possible before you begin the migration process in your data center.

5. Assess interdependencies

Part of the risk assessment phase (often referred to as the "discovery phase") should include a detailed assessment of the interdependencies between applications and IT equipment (servers, storage, and networks). This assessment will help determine the feasibility and details of a phased migration. It is recommended that organizations use a variety of tools during the risk assessment phase, including configuration management database (CMDB) tools.

6. Communicate

Executing an effective communication plan during all phases of a migration project is critical to eliminating rumors and false sources of information. Communication should involve all stakeholders and clearly assign roles and responsibilities.

Educating internal IT staff and affected business units about data center migration considerations is also key to avoiding surprises. Your HR can be of great help, especially if people are affected. Educate them about the significance of the migration, failures, and successes. Develop a detailed employee and management communication plan and provide regular updates (ideally via a website).

7. Planning

Data center migrations are typically conducted in phases. The number of phases often varies based on factors such as data center size, service risk level, budget and time constraints. Most organizations use a combination of the following multi-phase approaches:

First, organize the migration of low-risk parts and correct any problems encountered during the migration process in a timely manner.

Once the entire process is solidified, employees need to be well-trained in handling unexpected events before the organization migrates the high-risk parts.

8. Contingency Planning

Problems will arise during the migration process. The challenge for organizations is to detect these problems in advance and develop appropriate risk mitigation measures. The key to its success is to have a good and solid preparation phase. For example, a detailed inventory of IT equipment and network links is usually the basis of the contingency plan. In addition, temporary equipment and backup systems should also be included in the contingency plan.

9. Pre-test

Practice improved performance. Before migrating equipment, run a comprehensive set of tests to establish a "baseline" of operability, functionality, and performance of your infrastructure and applications.

10. Migration

To avoid unexpected errors, impose a "change freeze" period before the datacenter migration begins and ends after the migration.

During the migration process, be prepared to face common issues such as network connectivity issues, incorrect credentials (username and password), and missing validation/testing. These issues can have a cascading impact on other workload moves, especially when the upgrade path has not been properly determined. As a result, these issues can extend the freeze period and negatively impact the project.

11. Testing

For all testing, adopt a risk-based approach that incorporates the results of the Business Impact Analysis (BIA) and input from business subject matter experts (SMEs). For example, when performing application-level testing, ensure that the application owner/business unit is involved in the validation work and sign-off process.

Don't forget to include failover scenarios in your test plan.

12. Test after migration

Use the same test cases that you performed during the pre-migration phase and compare the post-migration results with the baseline results. Differences in results may indicate new issues that emerged during the migration process. Resolve any issues that are found.

After the migration is successful, pay special attention to the differences in online transaction processing performance, peak performance, and batch performance. Finally, carefully check whether all services have been fully tested in terms of functionality, elasticity, and performance.

13. Audit

It is recommended that organizations conduct a post-mortem review and audit of the migration project. The review should include an assessment of the following:

• Project Process
• Meets original business plan and design specifications
• In line with project schedule
• Feedback from project members and stakeholders

This review will provide the organization with valuable insights, critical success factors, lessons learned, and knowledge that can be incorporated into other projects within the organization. The lessons learned by the organization can make future projects more effective.

14. Close

Properly executing closure measures ensures that future hidden costs or expenses do not continue to impact IT operations. For example, take the following steps:

• Close the service contract for the original IT equipment
• Closing software contracts no longer in use
• Digestion of remaining storage assets (including copiers)
• Cancellation of contracts with suppliers of building management, maintenance services and supply services

15. Updates

Finally, after the data center migration is completed, the organization should update its processes, procedures, and documentation (including disaster recovery plans, compliance testing, and audit certificates). In addition, verify that all supporting systems (e.g., CMDB) have been updated correctly. The new data center may have a different setup, which may also require training for the operations team. Finally, recognition and marketing are also important. Through all efforts, promote the successful migration of the data center.