Resilient mobile core: rapid hybrid scaling with AI control

When mobile core capacity is needed, it is usually needed immediately. A stadium fills up, a regional outage redirects traffic, a maintenance window shifts demand, or an emergency event creates a sudden signaling surge. In these moments, customer experience depends on how fast the core can absorb new load before congestion, failed registrations or service degradation appear.

Reliability remains table stakes, as telecom networks are designed to achieve 99.999% uptime through carrier-grade engineering, redundancy, geo-redundancy and rigorous operational practices. However, recent outage scenarios show that reliability alone is not enough when failures are caused by software or procedural errors, malicious actors, massive power outages or other extreme conditions. Telecommunication providers therefore need to enhance their core networks by adding resilience on top of carrier-grade reliability, ensuring service continuity when conventional protection mechanisms are insufficient. Geo-redundancy and overprovisioning remain important, but they have economic limits; resilience requires the ability to adapt capacity and sustain service dynamically when the network is under stress.

That is why resilience can no longer depend only on permanently overprovisioned private infrastructure or manual recovery procedures. The mobile core needs an operating model that can expand capacity on demand, use hybrid cloud resources when they are needed, and protect critical core functions during abrupt traffic transitions. Nokia’s resilient mobile core approach combines cloud-native mobile core and voice core capabilities, intent-based automation and the packet core’s AI-assisted control to make that model practical for telecommunication providers.

The resilience challenge: capacity must be available before customers feel the impact

Traffic surges are rarely convenient. They can be triggered by emergency situations, major venues, regional outages, planned maintenance, or unexpected subscriber behavior. In each case, the operational requirement is the same: add capacity fast enough to prevent congestion, registration storms and service degradation.

Traditional approaches often rely on dedicated spare capacity, lengthy infrastructure preparation or manual workflows. That can increase cost and slow response when the network needs to adapt quickly. A hybrid scaling model changes the equation by keeping a warm standby mobile core footprint in public cloud at minimum capacity, then rapidly expanding it in predefined steps when additional capacity is required, making it ready to receive redirected traffic.

Rapid capacity augmentation without permanent overprovisioning

The solution enables telecommunication providers to augment mobile core and voice core capacity by using hyperscaler infrastructure outside the operator’s normal perimeter. Instead of investing upfront in permanent reserve capacity, operators can use a consumption-based model where additional licensing and resources are applied when the capacity is actually used.

This is particularly valuable for scenarios where demand is intense but temporary: large events, maintenance offload, disaster prevention, disaster recovery and controlled testing. Capacity can be scaled from a minimal standby environment to support millions of subscribers in minutes, reducing the need for long-lived idle infrastructure while improving operational readiness.

Automation to ensure rapid mobile core scaling

At the center of the architecture is Nokia automation, executed through NCOM (Nokia Cloud Operations Manager), to coordinate infrastructure and cloud-native network function scaling. The automation manages hyperscaler resources and CNF scaling as one integrated workflow, replacing fragmented manual procedures with standardized, repeatable operations.

This matters because hybrid cloud scaling across private and public clouds is not just an infrastructure task but relies on automation to rapidly sequence the events. For mobile core operations, the infrastructure, network functions, traffic steering and capacity policies must work together to move the RAN traffic from the original core to the hyperscaler based core. The architecture starts with a warm standby deployment in public cloud, sized to minimize cost while remaining ready for activation. When additional capacity is required, Nokia automation scales the hyperscaler infrastructure and CNFs in parallel, while traffic is redirected through mechanisms such as IP address switching at the radio network.  eNBs and gNBs traffic can be fully or only partially redirected, depending on network needs.

As demand changes, bidirectional traffic management supports both scale-out and scale-in. This allows the augmented environment to absorb additional load when required and return to a lower operating cost when the event or risk period has passed.

AI control protects the core during traffic transitions

Rapid scaling is only part of the answer. When traffic is redirected to the hyperscaler based core, the control plane can experience sudden peaks in registrations and service requests. Without protection, these peaks can affect core functions such as AMF (Access and Mobility Management Function), SMF (Session Management Function) and UDM (Unified Data Management), creating the risk of overload just when continuity is most important.

Nokia applies AI/ML capabilities in the packet core to monitor these conditions, detect abnormal behavior and control registration rates during traffic movements. By adapting anomaly models to the registration storm problem, the solution adds an intelligent protection layer that helps preserve service continuity while the network scales.

From resilience promise to operational advantage

For telecommunication providers, the value is not simply having more capacity available. It is the ability to turn resilience into an operating advantage: responding faster to demand spikes, reducing the cost of idle standby resources, and maintaining service continuity while traffic is being redirected. This changes resilience from a static design principle into a dynamic capability that can be activated when the network needs it most.

That shift is central to telecommunication providers’ evolution of their mobile cores. Cloud-native architecture provides deployment flexibility, hybrid cloud adds elastic capacity, automation coordinates scaling across infrastructure and network functions, and AI/ML helps protect the control plane during abrupt traffic transitions. Together, these capabilities give operators a more practical path to resilience without depending on permanent overbuild.

A more adaptive foundation for network resilience

The strategic question for telecommunication providers is no longer whether the mobile core can be made resilient, but how resilience can become part of the operating model. Networks must be ready to absorb disruption, support unpredictable demand and extend capacity beyond traditional boundaries without locking operators into permanent overbuild.

Nokia’s approach combines cloud-native core technology, hybrid cloud elasticity, automation and AI-assisted control to make that shift practical. It gives operators a path to build a mobile core that is not only more resilient, but more adaptive, more efficient and better prepared for the service demands ahead.