Dr. Somy Varghese, head of digital transformation and technology at a UAE-based luxury retailer, explores the shifting reality of digital resilience in an era where cloud infrastructure is deeply intertwined with geopolitical and physical risks.
For much of the past decade, the move to hyperscale cloud platforms has been viewed as the natural next step in digital transformation. Organisations migrated critical systems to the cloud with the belief that resilience would improve almost automatically.
Multiple availability zones, automated failover, and globally distributed infrastructure promised levels of uptime and operational continuity that few enterprises could realistically build themselves.
In many respects, that promise has been fulfilled. Cloud platforms have enabled organisations to scale rapidly, launch digital services with greater speed, and operate systems with impressive reliability. Entire industries—from banking and fintech to retail and logistics—have re-architected their digital operations around cloud platforms.
Yet recent disruptions affecting cloud infrastructure in the Gulf have prompted digital leaders to re-examine some of the assumptions behind that confidence. When drone strikes disrupted two availability zones within a regional cloud environment, the consequences were felt far beyond the data centres themselves.
Enterprise applications hosted in that region—platforms supporting customer engagement, operational workflows, analytics, and internal decision-making—suddenly became unavailable or severely degraded.
What surprised many organisations was not simply the disruption itself, but the time required for recovery. In several cases, systems expected to be restored quickly through backup mechanisms took days to return to normal service.
The episode exposed something rarely discussed in conversations about the cloud: the difference between theoretical resilience and operational recovery under real-world conditions.
The physical reality behind the cloud
Cloud computing is often described in abstract terms. Infrastructure appears to exist in a virtual space where resources scale automatically and services remain constantly available.
But beneath this abstraction sits one of the most complex physical infrastructures ever built. Hyperscale data centres are vast industrial facilities containing thousands of servers, power systems, cooling infrastructure, and high-capacity network connections.
These environments rely on stable energy supplies, telecommunications infrastructure, and regional logistics networks.
In strategic terms, cloud infrastructure has quietly become part of a nation’s critical infrastructure fabric. Financial systems, digital commerce, government services, and communication platforms all depend on these facilities.
The recent disruption highlighted an important reality: redundancy within a region does not eliminate exposure to regional events. When two out of three availability zones were affected simultaneously, systems designed to tolerate isolated failures suddenly faced a much broader disruption scenario.
For digital leaders, this is forcing a shift in thinking. Cloud resilience cannot be assessed purely through architectural diagrams or uptime statistics. It must also consider the geographic concentration of infrastructure, the geopolitical environment in which it operates, and the dependencies surrounding it. Digital resilience, in other words, is no longer purely a technical matter—it is increasingly a strategic one.
The hidden complexity of enterprise platforms
Another issue surfaced during the disruption: how little visibility organisations often have into the architecture of the enterprise platforms they depend on.
Most businesses interact with these platforms through service commitments such as uptime guarantees and service-level agreements. Beyond those metrics, the operational mechanics of these systems remain largely invisible.
Where backup environments are hosted, how disaster recovery systems are structured, and how quickly platforms can realistically be restored are details rarely visible to customers.
During normal operations this lack of transparency is rarely questioned. But during a disruption it becomes critical.
Recovery often involves multiple layers of providers: the enterprise software vendor, the cloud infrastructure operator, network providers, data replication systems, and distributed support teams across regions. Each layer operates with its own processes, escalation protocols, and responsibilities.
When something fails at scale, restoring services requires coordination across all these layers. Disconnects between vendors, shared infrastructure dependencies, and communication gaps can significantly slow recovery—even when backup systems exist.
The recent outages demonstrated how complex digital ecosystems can struggle to recover quickly when several providers are involved.
The risk of shared dependencies
The disruption also revealed how interconnected modern digital economies have become.
Banks, fintech firms, retailers, and service providers often rely on the same enterprise platforms and the same regional cloud environments. When those systems stopped responding, the effects cascaded rapidly across industries.
Customer engagement systems stalled. Operational dashboards went dark. Sales, service, and analytics teams lost visibility into key business functions.
In some organisations, teams discovered that multiple business-critical processes depended on the same underlying platform without fully recognising it as a strategic risk. In today’s digital economy, shared infrastructure creates shared exposure.
Multi-region architecture moves to the forefront
In response, many organisations are re-evaluating how their digital systems are designed. Historically, redundancy within a single cloud region—distributed across several availability zones—was considered sufficient for most enterprise workloads. This model balanced reliability with cost efficiency and worked well for common technical failures.
But disruptions affecting multiple facilities in the same region challenge that assumption.
As a result, organisations are increasingly exploring multi-region architectures where applications operate simultaneously across geographically separate locations. In these active-active environments, traffic can shift automatically if one region becomes unavailable.
For mission-critical systems, this approach is gradually evolving from a sophisticated architecture to a practical baseline for resilience.
Balancing resilience with reality
Of course, distributed infrastructure introduces additional cost and operational complexity. Running applications across multiple regions requires duplicated resources, cross-region data replication, and more advanced operational oversight.
Rather than applying maximum redundancy everywhere, many organisations are adopting a more strategic approach. Systems that directly impact revenue, customer experience, or regulatory compliance receive the highest level of resilience. Less critical workloads rely on simpler backup or recovery models.
When framed in terms of operational risk rather than infrastructure cost, the investment in resilience becomes easier to justify.
A more grounded view of digital resilience
Cloud platforms remain central to modern digital business. The cloud offers extraordinary reliability, yet it still operates within the physical and geopolitical realities of the world around it.
In this environment, resilience can no longer be measured purely by uptime statistics. It requires a deeper understanding of infrastructure dependencies, greater transparency from providers, and architectures designed to withstand disruptions that extend beyond purely technical failures.
As digital systems become ever more critical to economic activity, resilience is no longer just a feature of technology—it is becoming a core element of strategic risk management.
