Data Centre Decommissioning Project Plan in the UAE
By Ali Hassan, Editor, MAH Recycling
Most decommissioning projects slip at the start, not the end. Racks leave on schedule only when someone has already proven what is in them, what depends on them and who signs when they go.
This data centre decommissioning project plan for the UAE splits the work into eight stages, so a project manager can assign an owner and a sign-off to each one.
The eight stages at a glance
| # | Stage | What happens | Sign-off output |
|---|---|---|---|
| 1 | Scope and dependencies | Network scans, application owner interviews | Dependency map |
| 2 | Rack audit | Serial-level inventory, checked against discovery | Signed asset register |
| 3 | Freeze and runbook | Change-window freeze, shutdown order agreed | Approved runbook |
| 4 | Data decisions | Migrate, keep, purge or destroy, per media type | Data handling matrix |
| 5 | Power-down | Controlled shutdown, optional minimal-systems period | Shutdown log |
| 6 | De-racking | Top to bottom, cables removed and labelled | Handover sheet |
| 7 | Drive handling | Wipe or destroy per matrix, recorded by serial | Destruction records |
| 8 | Reconciliation | Records matched to register, buyback settled | Final report |
Stages 1 to 3: prove what you have
TechTarget’s decommissioning guide recommends network scans to find live systems, interviews with application owners to uncover dependencies, and a timeline with milestones and contingencies. It also advises comparing discovered assets with your established inventory, because the gaps between them are where surprises hide.
Treat TIA-942 as a baseline for the as-built room, not a decommissioning checklist. ANSI/TIA-942-C, published in May 2024, covers data centre infrastructure such as power, cooling and cabling. Its rating level tells you which systems your facilities team must safe-off.
Freeze changes before the audit finishes. A rack that gets a new VM halfway through is a rack you counted wrongly.
Stages 4 and 5: decide the data rule before power-down
Write a data handling matrix per media type before anyone touches a switch. Spinning disks, SSDs and NVMe drives need different treatment, and policy decides which leave intact.
Do not budget for multi-pass overwrites. NIST’s Revision 2 FAQ says they add nothing on modern media. For flash, plan on purge or destruction.
TechTarget also describes a “pilot light” period, where minimal systems stay running for roughly 30 to 60 days to catch overlooked assets.
Stages 6 and 7: de-racking and drive handling in the UAE
Remove equipment from the top of each rack downward for stability, and arrange loading docks, trolleys and forklifts beforehand. Cable removal takes longer than most plans allow: overhead trays, underfloor runs and wall-mounted management all have to come out.
In Dubai and Sharjah, building rules often set loading-bay hours and permit requirements. MAH Recycling agrees collection windows around those rules, and the loading-bay handover is where your serial-level register meets the buyer’s. Both sides sign it.
If policy says drives cannot leave your premises, our on-site HDD shredding service covers hard disks.
Stage 8: reconcile and settle
Match destruction records and the intake count against your asset register, resolve every difference in writing, then issue the final report.
Reusable servers, storage and network gear can be bought back. TechTarget’s rough guide says servers under two years old resell strongly, four-to-six-year-old servers still have potential, and anything older than seven years is mostly recycling value. Our server and network buyback page explains how we assess it, and settlement terms sit in the written quotation.
Send your rack list and building access rules through our data centre decommissioning page for a written scope, or browse all MAH Recycling services.
