Russia Reserves 10% of Private Data Center Capacity for State IT Needs

Russia plans to keep a slice of privately owned data center capacity ready for state use, creating a new pool of compute that could be called up on short notice. President Vladimir Putin has instructed the Digital Development Ministry to inventory capacity and floor space at private facilities. Owners could then be ordered to make the reserved resources available within 24 hours, Reuters reports.

The proposal sets aside 10% of capacity at non-state data centers. A newly formed commission would be able to issue the release order. The stated purpose is to give state IT infrastructure access to additional computing resources when needed. The source does not specify how operators would be compensated, how long a requisition could last, or which workloads would take priority.

Capacity on standby

For data center operators, the practical challenge is that capacity means more than racks with servers installed. Compute has to be powered, cooled, connected to networks and supported by staff. Space that looks available on an inventory may still need equipment or configuration before it can run a new workload. The source gives no details on how Russia’s inventory will account for those constraints.

A 24-hour compliance window leaves little time for ordinary planning. Operators would need a clear process for identifying which systems can be handed over, confirming that the hardware is ready, and coordinating access. They may also need to consider what happens to existing customers if reserved capacity is occupied or reassigned. The proposal, as described, does not explain how those operational questions would be handled.

Setting aside a fixed share can provide a straightforward measure of potential capacity, but it does not guarantee that every reserved server can run every application. Workloads depend on specific hardware, storage, network access and software environments. If the state needs a particular configuration, the usable pool could be smaller than the headline 10% suggests. That makes the planned inventory as important as the reservation itself.

Pressure on critical services

The plan follows damage to Russian data center infrastructure in recent weeks. In early October, drones struck the country’s largest data center, a 63-megawatt site. Multiple modules were taken out of service. Users reported trouble with Yandex Mail, Disk, Documents, Pay and Market, though Yandex said key services remained operational.

About two weeks later, another facility, used by Russia’s finance ministry, was damaged. It reportedly housed tax databases and other state registry data. Those incidents show why a reserve of distributed capacity may look attractive to service planners: if a facility loses equipment, another site could offer room to restore workloads. But transferring a service takes more than finding empty server space. Data must be available, systems must be compatible, and network connections must support the move.

That distinction matters for services tied to large, sensitive databases. A spare server pool can help with recovery only if the data and applications can reach it in time. The source does not say whether the reserved capacity will be geographically distributed, whether it will be kept in a ready-to-run state, or how data would be replicated. Those details will determine whether the reserve functions as a practical recovery option or mainly as a capacity ledger.

More than physical damage

Physical disruption is only one pressure on the country’s digital infrastructure. On the same day the capacity plan became public, the Russian Federal State Statistics Service acknowledged that its servers had been hit by cyberattacks. Reports said 23TB of data was stolen, including material about Russian losses in the conflict marked “for official use only.” The source does not connect those attacks to the proposed capacity reserve, but the timing highlights the range of risks facing data systems.

The report also says Russia damaged Ukrainian data centers in September, disrupting internet access for 100,000 households. A Ukrainian official described the network as highly decentralized and warned of local service interruptions. That example points to a basic design trade-off: spreading workloads across many sites can limit the impact of a single outage, while also making coordination and recovery more complex.

Russia has also floated temporary use of data centers outside its borders, subject to approval from the Federal Security Service. That option raises practical questions about latency, data transfer and the compatibility of services across locations. The report provides no details on which external facilities might be considered or what types of workloads could move.

For private operators, the immediate unknown is how the 10% reserve will be measured and maintained. For service users, the key question is whether capacity can be brought online quickly enough to protect the systems they depend on. The inventory, commission’s procedures and technical requirements will show whether the plan can deliver usable backup compute on a one-day deadline.

Reuters reports on the capacity order.

• The data center cooling state of play
• The custom AI ASIC state of play
• America’s AI chip rules keep changing — and the rest of the world is paying the price
• GTC 2026: Ian Buck press Q&A transcript — VP of Hyperscale and HPC speaks out on shelving CPX and shipping LPU decode this year
• Demand for data center CPUs has surged, and AI agents are responsible