Data Center Consolidation Strategy: Creating an Effective Project Plan

What do you see and hear when you look at racks of servers in a data center? You hear a lot of humming and see rows of blinking lights, and probably imagine that these machines are all hard at work. You’d be mistaken.

Historically, enterprise servers ran at only 5-15% utilization. Even today, many environments still operate far below optimal efficiency—often in the 20-40% range—despite gains from virtualization and containerization. This inefficiency is costly on many levels, and businesses and public sector organizations are taking notice. The US Federal government, for example, discovered that it could save over $600 million per year by consolidating data centers.

That’s why it’s important to consider why it may be wise to consolidate data centers and how to develop a meaningful data center consolidation strategy.

What is Data Center Consolidation?

Data center consolidation is the process of reducing an organization’s number of hosting facilities and migrating systems onto fewer physical hosts.

Consolidation can also include logical consolidation, such as increasing virtualization density, reducing the number of applications in the environment, or shifting selected workloads to cloud or SaaS platforms to eliminate redundant infrastructure. It’s about far more than just switching off servers and porting data from one data center to another. It’s a highly complex, multi-layered process that involves creating more efficient systems that are easier to administer and secure.

Reasons for embarking on a data center consolidation project include improving hardware and facility utilization, building more resilient infrastructure, and cutting costs. The process can also yield positive results for user experience and sustainability.

What is a Data Center Consolidation Strategy?

A data center consolidation strategy is a set of guiding principles that shape a data center consolidation project plan.

It should align with broader business goals while clearly articulating the desired outcomes of the process; for example, decreased energy use or optimizing network performance. At the same time, the strategy should provide a foundation for the organization’s IT future.

Examples of data center consolidation strategies include:

• Merge the systems running in four data centers into two while ensuring scalability and performance capacity to enable an artificial intelligence growth plan.
• Cut power and cooling expense by migrating systems from two data centers into one, while deploying high-performing hardware that aligns with anticipated workloads.
• Consolidate system administration teams, i.e., headcount, as part of a process of closing a data center.

Importance of Outlining a Data Center Consolidation Approach

Data center consolidation should not be an impulsive process. Doing it right means careful analysis, planning, and execution. The process will involve the participation of people from across the organization, as well as from external entities like co-location providers and consultants. Adopting the right data center consolidation approach is critical for success.

At the earliest stage in the process, before you even decide to consolidate data centers, it’s smart to gather the core stakeholders and try to figure out if consolidation is necessary, what its basic goals might be, and what, specifically, the project will involve.

What will the approach be? Will you start in small increments, migrating selected applications and data repositories to new hosting facilities, or will you plan a massive consolidation that takes a year to execute—completed with a big “cut over” moment? There’s no right answer, but it’s useful to think through your approach before you even start planning.

9 Steps of a Data Center Consolidation Project Plan

A data center consolidation plan should reflect deep thinking, analysis, and methodical execution of the data center consolidation process. Here are six project plan steps you can follow as you embark on your data center consolidation.

1. Define your Data Center Consolidation Goals

You can’t plan anything unless you know what you want to accomplish. This may sound obvious, but the scale and complexity of the task lead many to pursue “easy wins” that miss the crucial bigger picture.

Defining goals is a foundational planning step. A good practice is to start with high-level goals and figure out subsidiary goals that will enable their realization. For example:

• High-Level Goal: Cut data center operating costs
• Subsidiary Goals:
• Fewer sites in operation
• Reduction in the number of software licenses
• Reduction in administrative headcount
• Fewer physical servers deployed

• High-Level Goal: Improve resilience
• Subsidiary Goals:
• Simplified system architecture
• Consolidate BCDR systems and processes

• High-Level Goal: Greater scalability
• Subsidiary Goals:
• Optimal virtualization to ensure room to grow
• Optimal hardware for future growth and performance needs

Also important is to establish goals relating to the scope of consolidation. For example, a data center consolidation project may be outlined to consolidate the infrastructure hosting the Enterprise Resource Planning (ERP) solution. However, the end result may not be an actual data center consolidation, but rather the migration of the ERP from hosting at three sites to one. This will realize goals such as cost savings and better BCDR—while laying the groundwork for future shutdowns of the data centers where ERP was previously hosted.

2. Create a Data Center Consolidation Checklist

A data center consolidation checklist provides an efficient mechanism for planning and communication amongst diverse stakeholders. The specifics will vary by organization and project, but a good checklist will include the following items:

1. Conduct an extensive equipment, software, and facilities inventory.
2. Discover all applications and dependencies between systems. This can include determining application performance baselines and the metrics you would want to improve.
3. Build a project team with executive sponsorship and representation for all relevant stakeholders, e.g., finance, legal, facilities, security, compliance teams.
4. Establish a mode of communication for all parties involved.
5. Develop a framework for consolidation that highlights potential risks while maximizing potential beneficial outcomes. For example, some workloads are better suited for on-prem vs colocation vs private cloud vs public cloud. Transitioning them away from their natural environment could be damaging.
6. Thoroughly analyze the total cost of ownership (TCO) that takes into account business objectives for the consolidation.
7. Develop and implement an ITAD process for decommissioning, disposing of, and recycling hardware.
8. Chart system/application configuration and dependencies with the goal of maximizing agility and resiliency in the future.
9. Develop a comprehensive risk framework accompanied by clear, proactive mitigation processes.
10. Integrate a BCDR plan that encompasses old and new facilities—and test it.
11. Determine reasonable and permitted downtime for the consolidation process.

If a checklist like the above is followed, the project has a greater chance of success with minimal downtime.

3. Conduct IT Asset Discovery

Thorough discovery is the basis for a successful consolidation. If you don’t know what you’re running in a data center, your consolidation is headed for trouble. You need to catalogue servers, storage systems, networking hardware, and software-defined assets. You also need to know where all these systems store their data.

For example, an ERP solution may function through integrations with Customer Relationship Management (CRM) solutions, logistics systems, sales management, financial systems and legacy databases. Moving such technology can be a huge challenge.

The first step in avoiding problems here is to gain a complete understanding of where the integrations are and how a migration of a data center to a new hosting site will affect them.

Separately, the discovery process needs to document energy use and the physical aspects of the target environment. If you are running equipment on 100 52U racks, and you know you can shrink the physical footprint down to 80 in the new facility, you should confirm that there will be space for those 80 racks.

Automated discovery tools are the standard method for conducting this process. This will likely lead to identification of underperforming IT assets such as “zombie servers” that are running without any performance benefits, while using energy and cooling capacity.

4. Map Dependencies

Dependency mapping shows how all the various equipment items in the IT infrastructure connect to one another. It takes discovery to a new level.

The goal is to document relationships between applications, workloads, data, and the infrastructure that supports them. This might mean, for instance, detailing how a single sign-on (SSO) server enables login to a group of systems. Knowing these dependencies and relationships is essential for developing a data center relocation plan that will result in relevant systems working as intended once they move to a new site. This mapping should also account for micro-segmentation rules, firewall ACLs, Zero Trust access enforcement points, and east–west traffic patterns, all of which must be replicated or redesigned during consolidation.

One tactic that can be useful here is to examine past data center outages and determine if any points of integration are particular sources of risk. For example, if one API is at the root of three outages, that can tell you something about how important it will be to make sure that it migrates properly to its new home.

4b. Map Your Network

A subsidiary of mapping dependencies is mapping the network of the entire infrastructure. This should be dynamic, holistic (and incorporate existing cloud environments), and encompassing of all entities (including remote locations, companies acquired through M&A, and global sites).

Modern network mapping should also include Layer 2/3 topology, routing policies, DNS/DHCP dependencies, SD-WAN pathing, and any cloud interconnects (for example, Direct Connect or ExpressRoute). These have direct impact on migration sequencing and cutover success.

Realistically, this can only be achieved using best of breed network mapping software that allows complete and easy visualization of network inventory – including assets, incidents, severity and likely ramifications. Mapping will form a multi-layered map that consolidates server information into a single server topology globally showing network paths, services, and instances in one view.

5. Envision the Future State and Plan for its Realization

A consolidation project needs to feature a complete vision of the future state and a plan for getting there. Keep in mind that the ideal goal is to not just replicate what you have in the old data center. The consolidation is a great opportunity to do things better. For example, if slow arrays are hampering the server performance at the old site, then it is important to upgrade the performance of these server arrays.

Improvements might also include a consolidation of Data Center Infrastructure Management (DCIM) systems, or staffing configurations. Many organizations may even use consolidation as an opportunity to modernize infrastructure through hyperconverged infrastructure (HCI), software-defined networking (SDN), composable infrastructure, or private cloud platforms. These can significantly improve scalability, automation, and lifecycle management.

It could be an opportunity to also figure out how to improve power and IT infrastructure cooling characteristics; for example, implement liquid cooling for certain workloads. Execution matters, though. You can plan for your dream data center, but you need to map out a clear plan of execution.

6. Undertake Migration Planning and Contingencies

Once the ideal state of the consolidation has been projected, you need to develop a thorough plan with built-in contingencies. Most aspects of enterprises today depend on exhaustive server uptime, so any blip in service can have a dramatic impact.

Before addressing contingencies, define your migration and cutover strategy. Options include big-bang cutover, phased migration, parallel-run, or blue/green deployment.

For data movement, you’ll need to select the proper migration method (block-level replication, database replication, storage vMotion-style migrations, or application-level exports) depending on workload sensitivity, downtime limits, and data consistency needs.

You need to be ready for unexpected events, such as:

• Incompatible legacy software and new hardware
• Supply issues for power and cooling at the target location
• Lower than expected levels of security during transit and setup
• A skills gap for an enterprise-level data center consolidation project

As with all change management, practices and reviews should be conducted frequently and rehearsed prior to migration rollout. Having a disaster recovery plan (DRP) in place before execution is essential.

Consolidation often changes network distances and routing, so it’s critical to evaluate WAN capacity, latency impacts on remote offices, SD-WAN routing, and any MPLS circuit changes. These factors can materially affect application performance after consolidation.

7. Decide on Architecture and Design

The design of your consolidated data center will ultimately depend on your business requirements. Depending on your tech stack, you will not have to start from scratch. Many of the leading OEMs provide reference architectures you can leverage to build out your consolidated site.

Depending on your budget, modular data center layouts exist that you can leverage within your facility. This may even help you scale up and down within your new consolidated facility when your utilization and capacity needs change.

8. Data Center Build

Once your consolidation plan is ready and your contingencies have been explored, you can assemble your execution team. At a minimum, your team needs to account for IT teardown and setup, logistics, procurement, and contracting professionals.

If you’re engaging in any new construction, the project may require extensive outside help for construction, legal counsel, real estate guidance, project management, energy consulting, and so forth.

Regardless of size, it will be important to understand the scope and responsibilities of involved parties, the dispute resolution process, and what to do if timelines aren’t met.

The build phase should also account for hardware lifecycle considerations, including OEM/EOSL timelines, firmware alignment across hosts, storage compatibility lists, and hypervisor support matrices. These technical constraints often affect which systems can migrate and when.

9. Execute and Plan for Continuous Improvement

The key to success with the data center consolidation methodology is to understand how you’re going to implement each stage of the consolidation plan, including testing for problems.

You will have thought through the steps, what will constitute success, and how to check if it’s working the way you want it to. For example, if the plan calls for migrating a database, you have to test that you’ve transmitted all the data and that it writes onto the new storage equipment without any corruption. You’ll also need to validate that any security controls and compliance measures are in place.

The execution process does not stop at cutover, either. Inevitably, you will have to do extensive tuning and adjusting of configurations to reach your stated goals.

From there, you will ideally pursue continuous improvement. As part of this process, it is probably smart to deploy a tool that tracks utilization, hardware performance monitoring, security posture, and Power Usage Effectiveness (PUE) on an ongoing basis. This tool will provide the metrics you need to track how well your consolidation is going and help you figure out how to do better in the future.

Data Center Consolidation Project Plan Template

One effective tool in facilitating a successful data center consolidation is an effective project plan template. These vary widely in format and level of detail, but however you design it, the template offers an excellent way to map out the schedule, expectations of tasks, and their completion dates.

They’re a great way to communicate what the plan requires from a diverse set of stakeholders. They are also good for highlighting dependencies between tasks; for example, that you can’t perform the inventory until you’ve completed the risk assessment.

Here’s an example of a simple approach that can be built out further:

Considerations when Creating a Data Center Consolidation Plan

A data center consolidation project can turn into a massive undertaking. With the scope, schedule, and long list of stakeholders, it can be easy to overlook certain key considerations. As you contemplate doing a consolidation, here are some things to look for:

• How will data affect the future-state design? High-volume datasets must often remain close to the applications that use them, influencing which workloads can be consolidated.
• Does the consolidation introduce new compliance implications—for example, cross-border data residency, industry regulatory changes, or new encryption requirements?
• Where is your organization headed, strategically, and how will that affect the consolidation? For example, if there’s a customer-facing digital transformation in the works, with omni-channel communications and mobile apps, will that have an impact on the consolidation plan?
• What is your organization’s cloud strategy? Many of the systems you want to port over to a new data center might be better off in the cloud. With this in mind, cloud suitability assessments should determine whether workloads are refactored, rehosted, retained on-prem, or retired. Many consolidation projects reduce physical DCs while increasing cloud footprint.
• Are there evolving regulatory issues that could have an impact on the consolidation, e.g., if a major data privacy law is about to come into effect that changes the rules for data sovereignty, you have to factor that into the consolidation plan.
• Is your organization planning to consolidate vendors? IT Procurement is an under-appreciated aspect of consolidation. The process may offer an opportunity to go all-Dell or all-Linux to save money and streamline administration. Or, if you’re not considering vendor consolidation as part of the data center consolidation, perhaps you should.

Ease the Data Center Consolidation Process with Park Place Technologies

Following data center consolidation steps and engaging in merging infrastructures can seem like an insurmountable task carrying a high workload, which is why some companies shy away from the process. The benefits however are far reaching. To tap straight into the benefits and bypass consolidation adoption issues, it’s important to seek out a reputable global data center services firm.

Park Place Technologies offers data center hardware maintenance, IT professional services, infrastructure managed services, and enterprise IT infrastructure monitoring software–all from a single source. Eliminate the hassle of working with multiple vendors.

Our IT asset relocation services can also be on hand, to ease your IT infrastructure migration processes.

Let our team help you find efficiencies in your IT operations today!