CYBER RESILIENCE FOR LARGE-SCALE DATA CENTER MIGRATION IN SAUDI CRITICAL INFRASTRUCTURE UNDER VISION 2030: SECURE IP FABRIC SEGMENTATION, THREAT CONTAINMENT, AND OPERATIONAL CONTINUITY
Large-scale data center migration within Saudi critical infrastructure is not a conventional relocation of servers and applications. It is a time-bounded transformation of trust boundaries, routing domains, identity dependencies, recovery mechanisms, and operational accountability. During coexistence, legacy and target environments remain interconnected, which expands the attack surface precisely when configuration volume, change velocity, and service uncertainty are highest. This review examines how cyber resilience can be engineered into migration programmes through secure IP fabric segmentation, evidence-led threat containment, and continuity controls aligned with Saudi Vision 2030. An integrative review of peer-reviewed studies, standards, and regulatory documents published from 2020 to 2025 was undertaken. Evidence was synthesised around five analytical themes: migration risk, EVPN-VXLAN segmentation, zero-trust enforcement, containment and recovery, and governance. The review finds that resilient migration depends less on a single security product than on coordinated design decisions. These include separating management, replication, user, backup, security, and operational-technology flows; replacing inherited network trust with identity- and workload-aware policy; constraining migration corridors; maintaining cryptographically protected recovery copies; and releasing each migration wave only after observable technical and business evidence has been obtained. A reference operating model is proposed in which dual-running environments are governed through explicit security zones, continuous telemetry, policy-as-code, tested rollback, and service-level recovery objectives. The principal contribution is a practical review framework that treats migration as a sequence of reversible resilience decisions rather than a one-time cutover.