A data center rarely closes for just one reason. A lease may be expiring, a colocation contract may be ending, a corporate consolidation may be underway, or an owner may be retiring a legacy facility in favor of newer capacity built for higher density workloads. Whatever the cause, the moment a data center is designated for closure, the operator faces a project that touches IT infrastructure, electrical and mechanical systems, real estate obligations and, in many cases, sensitive data — all on a timeline usually dictated by a landlord, a corporate deadline or a service contract that doesn’t pause for logistics.
USA Decom provides full data center decommissioning services to facility owners, operators, corporate real estate teams and IT infrastructure groups throughout the United States. Because this work happens in markets across the country, our approach is built around the realities of multi-site, multi-state coordination: consistent project management, adaptable crews and equipment, and a process that holds up whether the facility is a single enterprise server room, a regional colocation deployment, or a large-scale data hall being retired ahead of a new build.
Why Data Centers Get Decommissioned
Several conditions typically trigger a full decommissioning project rather than a smaller equipment swap or partial upgrade. A company may be exiting a leased facility and returning it to the landlord in a specified condition. A corporate IT department may be consolidating multiple regional data centers into a centralized cloud or colocation strategy. An enterprise may be retiring an on-premises facility entirely as workloads shift off-site. In other cases, an aging facility reaches the point where the electrical and mechanical infrastructure can no longer support current density requirements, and replacing that infrastructure costs more than relocating operations to a purpose-built site.
Many older data centers were originally designed around lower power densities than current IT equipment requires, which puts operators in a position of choosing between a significant retrofit or retiring the facility in favor of newer capacity. That decision is highly specific to the facility, its location and the condition of its remaining infrastructure. Whatever drives it, once the decision is made, the operator needs a plan that gets the space emptied, the data protected, the assets accounted for and the facility turned over on schedule.
What a Complete Data Center Decommissioning Project Involves
A full decommissioning differs from a simple equipment pull or a white space clear-out. It’s a coordinated project that typically starts at the server and rack level and works outward through every supporting system in the facility. Depending on scope, a project can involve:
- Servers, storage arrays, network switches and other IT equipment
- Racks, cabinets, containment and structured cabling
- UPS systems and associated battery strings
- Standby and emergency generators
- PDUs, RPPs, busway and branch circuit distribution
- Switchgear, transformers and upstream electrical infrastructure
- CRAC and CRAH units and associated mechanical piping
- Raised flooring, fire suppression tie-ins and monitoring/BMS systems
Rather than treating each of these as an isolated removal task, we build a sequence that respects how the systems interact. Power has to be de-energized in the correct order, mechanical systems have to be isolated once the IT load they support is gone, and physical removal has to happen without disrupting adjacent tenants, ongoing operations elsewhere in the building, or the building’s own base infrastructure if the facility isn’t being demolished outright.
Planning and Scope Development
Every decommissioning project starts with understanding what’s actually in the facility and what condition it needs to be in when the work is finished. That means gathering an accurate inventory of IT and infrastructure assets, confirming utility service points and isolation locations, reviewing lease or purchase-agreement language that defines the required turnover condition, and identifying any equipment the customer intends to keep, relocate to another site, or sell. This planning stage also has to account for building access — loading dock capacity, freight elevator dimensions and weight limits, parking for roll-off containers and trucks, and any building management restrictions on work hours or noise.
For multi-site or multi-state decommissioning programs, planning also includes standardizing the process across locations so that reporting, data handling and asset disposition are handled consistently no matter which region the facility is in. Corporate customers managing several closures on the same timeline benefit from a single point of contact managing scope, schedule and documentation across every site rather than negotiating a separate process with a different contractor in each market.
Shutdown Sequencing and Active-Facility Coordination
Few data centers are decommissioned instantly. In many cases, some portion of the facility remains live — supporting a partial migration, a colocation tenant still in the process of moving out, or building systems that serve other parts of the property — while another portion is cleared. That means shutdown sequencing has to be planned circuit by circuit and system by system, confirming with the customer’s IT and facilities teams exactly which loads can be de-energized and when.
This is where close coordination with the customer’s engineering staff, electricians and, where applicable, the building owner becomes critical. Before any UPS, PDU, generator or switchgear component is disconnected, isolation points need to be verified and appropriate lock-out/tag-out procedures followed. Working in an occupied or partially occupied building adds another layer: dust, noise and equipment movement have to be managed so they don’t affect neighboring tenants or ongoing operations, and any freight or elevator use has to be scheduled around the building’s normal traffic.
Data-Bearing Equipment and Chain of Custody
Servers, storage arrays and networking equipment typically hold sensitive corporate, customer or regulated data, which makes this equipment different from the rest of the facility’s assets. Many organizations specify data sanitization standards such as the Clear, Purge and Destroy methods described in NIST 800-88, or an equivalent internal policy, as part of their decommissioning requirements. Data-bearing devices may require appropriate sanitization, destruction or chain-of-custody documentation as part of the overall decommissioning plan, and customers should identify their specific data-handling requirements — including any regulatory obligations tied to their industry — early in the scope development process so the removal sequence and documentation can be built around them from the start.
Maintaining a clear chain of custody matters not just for data security but for asset tracking generally. A facility being fully decommissioned may contain hundreds or thousands of individual components moving out through different disposition paths — some kept by the customer, some sold, some recycled, some destroyed — and a documented tracking process keeps that inventory accountable from the moment equipment is disconnected to the moment it leaves the site.
Electrical Isolation, Rigging and Physical Removal
Once systems are de-energized and cleared for removal, physical extraction becomes a rigging and logistics problem. UPS cabinets and battery strings are heavy and often located on upper floors or in rooms with limited clearance. Generators may be roof-mounted, in exterior enclosures, or in below-grade vaults that require specialized lifting and routing plans. Switchgear and transformers can weigh several thousand pounds and require calculated lift points, appropriate rigging hardware and a clear path to the loading area. Battery strings associated with UPS systems also require careful handling and appropriate transportation classification given their chemical composition.
Server racks and cabling are comparatively lighter but time-consuming, since structured cabling often needs to be traced, labeled or documented before removal, particularly if any portion of the facility will remain in operation. Raised floor systems, once cleared of underfloor cabling and cooling ductwork, may need to be dismantled and removed or left in place depending on the facility’s turnover requirements.
Asset Recovery, Recycling and Equipment Disposition
Not everything in a data center headed for decommissioning is scrap. Generators, UPS systems, PDUs, switchgear, transformers, CRAC units and even server hardware can retain resale or reuse value depending on manufacturer, capacity, age, condition and completeness. Before equipment is scrapped or destroyed, it’s worth evaluating what can be sold, reused elsewhere in the customer’s portfolio, or recycled for its commodity value in copper, aluminum, steel and other recoverable materials. Well-maintained infrastructure equipment — a Liebert or Vertiv UPS, a Caterpillar or Cummins standby generator, Eaton or Square D switchgear — can carry secondary-market demand, and identifying that value early can help offset a portion of the overall project cost, though it won’t necessarily cover the full scope.
Recycling and salvage planning also has an environmental dimension. Batteries, refrigerants from CRAC and chiller systems, and various electronic components require appropriate handling depending on applicable requirements, and separating recoverable materials from true waste reduces what ultimately goes to landfill. A decommissioning plan that accounts for disposition upfront, rather than treating everything as demolition debris, generally produces a cleaner, better-documented and more cost-effective outcome.
Site Restoration and Final Turnover
The last phase of a decommissioning project is bringing the space to whatever condition the lease, sale agreement or corporate requirement specifies. For a leased facility, that might mean returning the space to landlord standard — patching floor penetrations, removing raised flooring and containment, restoring ceiling grids, and clearing all data center-specific infrastructure back to base building conditions. For an owned facility being repurposed or sold, turnover requirements may be different, focused more on complete equipment removal and a clean, safe shell ready for the next use. Confirming these requirements before work begins avoids rework and disputes at the end of the project, and a final walkthrough with documentation of the completed condition gives the customer a clear record for their landlord, buyer or corporate stakeholders.
Real-World Project Examples
Data center decommissioning projects we’ve supported typically follow a similar arc regardless of region: an inventory and scope walk, a sequencing plan built around what can go offline and when, coordinated electrical isolation, systematic removal working from IT equipment outward to infrastructure systems, and a documented asset disposition process running throughout. On a typical enterprise data center closure, that can mean removing servers and networking devices with documented data destruction, followed by rack and cabinet teardown, then UPS and battery removal, generator disconnection and rigging, and finally CRAC unit and electrical infrastructure removal — sequenced so the space is progressively cleared without compromising any area still supporting live operations elsewhere in the building. On multi-location programs, the same process gets repeated at each site with consistent reporting so a corporate customer can track progress and asset disposition across every closure from one point of contact rather than piecing together updates from different regional contractors.
Coordinating a Nationwide Project
Because data center decommissioning projects happen wherever a company has infrastructure — corporate campuses, leased colocation suites, regional offices, and legacy facilities in secondary markets — a nationwide approach has to flex to different building types, local logistics and regional labor and disposal markets while keeping the underlying process consistent. That includes adapting to regional differences in scrap and recycling infrastructure, transportation distances for oversized equipment like generators and transformers, and varying building access conditions from downtown high-rises to suburban office parks and standalone data center campuses. USA Decom provides data center decommissioning services across the United States, applying the same planning discipline, documentation standards and asset disposition approach whether the project is a single facility or part of a broader portfolio consolidation.
What to Prepare Before Requesting a Project Review
Before reaching out, it helps to have a general sense of the facility’s size, the IT and infrastructure equipment involved, any known data-handling or regulatory requirements, the target completion date, and the condition the space needs to be in at turnover. Photographs of the data hall, electrical rooms and mechanical spaces are useful even at an early stage, and any existing equipment inventory or as-built documentation speeds up scope development considerably. None of this needs to be finalized before making contact — much of it gets refined during the planning process — but the more detail available upfront, the faster an accurate project review can come together.
Start Your Nationwide Decommissioning Project
If your organization is planning to close, consolidate or retire a data center anywhere in the United States, use the form below to tell our team about the facility type, location, approximate size, equipment involved, data-handling requirements and target schedule. Whether you’re managing a single closure or coordinating multiple sites on a corporate timeline, sharing these details lets us put together an accurate scope and sequencing plan for your project.
