Skip to content
Request an assessment
The WesternTechSystems facility facade at blue hour

Services · Data Center Decommissioning

A facility exit is a records project with freight attached

The equipment will move regardless. What separates a clean decommissioning from an expensive one is whether, eighteen months later, you can still say what every serial was, where it went and how it was sanitized.

The failure mode

Nobody loses a data center. They lose the record of one.

Decommissioning goes wrong in a predictable way. Discovery is compressed because the lease date is fixed, the inventory is taken from a register nobody has reconciled in three years, equipment is staged in a corridor, and the project generates evidence in fragments — a spreadsheet here, a certificate batch there, a recycler's weight ticket that references nothing.

Then, later, someone asks what happened to a specific storage array. And the honest answer is that it cannot be established.

That is not a logistics failure. It is a failure to treat the exit as a records exercise from the first day — which is why our sequence starts with a serialized opening inventory and ends with one closeout package built against it.

Seven phases

What happens, in what order, and what each phase leaves behind

Every phase produces a document. The closeout package is those documents assembled against a single opening baseline — not collected afterwards from whoever still has them.

  1. 01

    Discovery

    Serialized opening inventory built against your CMDB or asset register, with variance identified before anything moves. The inventory is the baseline every later reconciliation is measured against.

    Produces
    Opening inventory and variance report
  2. 02

    Decommissioning engineering

    Sequence, freight profile, packaging, custody model, legal-hold handling and the execution plan — authored by our technical services engineers and agreed in writing before a date is booked.

    Produces
    Decommissioning plan and scope statement
  3. 03

    On-site execution

    Our national field crews perform the physical decommissioning under our own project management — equipment removal and de-racking within the managed scope, labelling, packaging and staging — then palletized and manifested against the opening inventory. LTL or FTL by lane, with EDI 214 in-transit visibility rather than a tracking link somebody has to check.

    Produces
    Custody transfer, outbound manifest, in-transit visibility
  4. 04

    Secure receiving

    Inbound assets enter a mesh-enclosed, lockable receiving cage and are reconciled against the manifest before joining general stock. Legal-hold items segregate immediately.

    Produces
    Chain-of-custody receipt, reconciliation
  5. 05

    Sanitization

    Every data-bearing component sanitized against NIST 800-88 by media type, with a certificate issued per device. No asset advances without a valid certificate; manual overrides are not permitted.

    Produces
    Per-device erasure certificate
  6. 06

    Valuation & disposition

    Configuration-level identification, testing and grading. Equipment with a market is remarketed; the rest is consigned to qualified, certified downstream recyclers under manifest.

    Produces
    Settlement statement, downstream manifests
  7. 07

    Closeout

    The complete evidence package assembled as one record: reconciled inventory, certificates, custody transfers, settlement and downstream manifests.

    Produces
    Project closeout package

Why infrastructure is different

An endpoint process applied to a data center gets both halves wrong

A populated rack is many assets, not one line

A 2U server is a chassis, a set of drives, memory, processors and sometimes an accelerator, each with its own handling requirement and its own residual value. Providers who inventory at chassis level are simultaneously under-sanitizing and under-recovering, and neither error is visible until much later.

Storage is the actual risk

Arrays, caching tiers, boot media, RAID controllers with battery-backed cache, and drives that were pulled and replaced years ago and never left the room. Data-bearing media in a facility exit is rarely where the register says it is.

Network gear holds data too

Switches, routers, firewalls and load balancers carry configurations, routing tables, credentials and VPN parameters. A great deal of decommissioned network equipment leaves buildings unsanitized because nobody classified it as a data-bearing asset.

Depreciation runs on a clock

Infrastructure loses secondary-market value fast, and equipment that sits in a staging area for three months while a project stalls is worth measurably less than the same equipment moved in week one. Sequencing is a financial decision, not just a logistical one.

Scope

Where our scope starts, in writing, before a date is booked

Decommissioning RFPs frequently go wrong because two parties assume different answers to the same question. So we state the boundary explicitly rather than leaving it to the statement of work.

On your floor
Decommissioning engineering and planning, project scoping and execution planning, technical project management, and on-site technical execution by our own national field crews — including equipment de-racking and removal as part of the managed project. This is WesternTechSystems Technical Services, not a subcontracted introduction.
From your floor to ours
Custody transfer at the point of removal, manifested transport, and secure receiving into a mesh-enclosed cage reconciled against the opening inventory before anything joins general stock.
In our facility
Sanitization, testing, configuration-level valuation, remarketing and the closeout evidence package — performed in our own R2v3 and RIOS certified facility in Lewisville, Texas.
Beyond our facility
Material with no reuse or resale route is consigned to qualified, certified downstream recyclers under our R2v3 Appendix A downstream vendor management scope, with manifest documentation. The physical recycling is performed by those vendors, not by us — what we perform is the qualification, the routing and the documentation that makes the chain auditable past our loading dock.
Named specialist scope
Some facility exits require functions outside our scope — teardown of liquid-cooled systems, fluid handling, or specialist heavy rigging beyond managed project removal. Where a project needs one, it is identified during planning and the scope statement names who performs it. We would rather write that down before a date is booked than discover it on the morning of the exit.
What we will not imply
That a certificate covers more than it does. Our R2v3 data-sanitization appendix covers logical sanitization. Where your policy requires physical destruction of media, say so during scoping and we will tell you plainly how that is handled and by whom.

The control worth knowing about

Unsanitized assets cannot be shipped or invoiced

In our highest-volume programs, wipe validation is enforced at two separate points — at label creation and again in ERP fulfilment. An item without a passing wipe record is blocked from finished goods, from shipping and from invoicing.

Blocked, not flagged. The distinction matters more than it sounds. A flag depends on somebody reading it. A block does not.

The same architecture governs decommissioning work. Sanitization is not a step that precedes release; it is the condition of release, and the certificate is what satisfies it.

How sanitization works here

FAQ

What decommissioning project owners ask

Where does your scope start and stop?
It starts on your floor. WesternTechSystems Technical Services provides decommissioning engineering and planning, project scoping, technical project management, national technical field crews and on-site technical execution — including equipment de-racking and removal as part of the managed project. From there it runs through custody transfer, transport, secure receiving, reconciliation, sanitization, testing, valuation, remarketing, downstream recycling and closeout. Where a project requires a specialist function outside our scope, it is named in the same scope statement rather than left to assumption.
How do you handle a facility where the register is out of date?
Assume it is, and build the opening inventory from the physical estate rather than the document. Discovery exists precisely because CMDBs and reality diverge, and the variance between them is one of the more useful things a decommissioning project produces. Reconciling to a stale register just moves the problem into the closeout package.
What happens to drives?
Every data-bearing component is treated as an asset in its own right, sanitized by the appropriate method for its media type against NIST 800-88, and issued a certificate carrying serial, method, operator, station, timestamp and validation hash. Drives are not sanitized as a batch and certified as a lot.
Can equipment be sanitized before it leaves our site?
Raise it during planning. Some organizations require it as policy, and there are good reasons for that position. There are also trade-offs — on-site sanitization is slower, harder to verify independently and produces evidence generated outside a certified environment. We will give you the honest comparison for your situation rather than a default answer.
How is legal hold handled?
Held assets are identified at planning, flagged on the manifest, and segregated into restricted-access storage immediately on arrival. They cannot be sanitized or transferred until a written release is received and logged from your authorized representative. That is a system control, not a note on a spreadsheet.
What is retired infrastructure actually worth?
Server hardware typically recovers in the range of 5–20% of original cost and networking equipment 10–25%, based on our own published recovery benchmarks. Where a specific estate lands inside those ranges depends on generation, configuration, support status and condition — which is why valuation happens at configuration level rather than by chassis count. See asset recovery for how the ranges are derived.
Can we witness the process?
Client compliance officers and designated third-party auditors can schedule a facility walkthrough and records review with 48 hours' notice, during your project or at any other time. If witnessed processing of specific assets is a requirement, raise it at scoping so it can be built into the sequence.
How fast can you move on a lease deadline?
We deploy field crews nationally and our surge staffing supplements the core team within 24 to 48 hours, with dedicated processing zones stood up inside a day. In practice the constraint on a deadline-driven exit is almost never our processing capacity — it is discovery. The projects that run late are the ones where nobody knew what was in the room until the week the trucks were booked.
What do we receive at the end?
One closeout package: the reconciled inventory against your opening baseline, an erasure certificate per data-bearing device, documented custody transfers, the settlement statement for remarketed equipment, and downstream recycling manifests for material that could not be reused. Assembled as a single record rather than as attachments across nine months of email.
Do you operate nationally, or only around Dallas?
Nationally. WesternTechSystems Technical Services fields technical crews for on-site execution across the country, managed by our own technical project managers, with processing consolidated into the certified facility in Lewisville, Texas. Geographic reach is a scoping question rather than a limiting one — what varies by site is sequencing, access and freight lane, not whether we can be there.
What if the facility contains AI or accelerated computing hardware?
It is scoped into the same project rather than handled separately, and it changes the inventory and valuation work rather than the custody model. An accelerated computing chassis is a much longer list of components than a conventional server, with very different residual values across it. See GPU and AI infrastructure lifecycle management for how that estate is treated, including the parts of it we are explicit about not claiming.
Do you recycle the material yourselves?
No, and it is worth being precise about this because the industry is not. We are an R2v3 and RIOS certified processor: intake, sanitization, testing, repair, grading, packaging and remarketing happen in our building. Material with no reuse or resale route is consigned to qualified, certified downstream recyclers who perform the physical recycling. What we perform is the qualification of those vendors under R2v3 Appendix A, the routing decision, and the manifest documentation that lets the chain be audited past our loading dock. A responsible-recycling claim that stops at a provider's dock is not a claim about anything.
Do you handle a phased exit rather than a single event?
Yes, and phased exits are usually the right structure when production workloads are still migrating. The important thing is that each phase reconciles against the same opening inventory, so the final closeout is one continuous record rather than several partial ones that have to be joined afterwards.

Start with discovery, not with a date.

Tell us the site, the deadline and what you believe is in the room. The gap between that belief and reality is where the project actually lives.

Request an assessment (469) 718-0004