The Ground Beneath the Cloud: Dispersion and Dependency in the Digital Age
Cloud providers spread workloads across data centers so customers can keep using services when one site fails. But that protection depends on physical sites, power, networks, and local conditions. When damage spreads across those sites, shifting workloads cannot guarantee recovery. In September, AWS said it could no longer restore Bahrain-only resources and data after physical damage. Damage spread across multiple availability zones, beyond what its regional design was built to withstand. In the UAE, AWS made the same determination for one of three zones; recovery continued in the other two. Customers initially moved workloads to other regions, limiting the strike's effects, but physical damage eventually overwhelmed that protection. The episode shows moving workloads is no guarantee: other regions depend on shared networks, supplies, and host governments. Going forward, digital resilience requires more than spreading workloads; it also depends on protecting the infrastructure that connects them.
What happened to AWS services in Bahrain and the United Arab Emirates after physical damage hit their data centers?
Cloud providers spread workloads across data centers so customers can keep using services when one site fails. But that protection depends on physical sites, power, networks, and local conditions. When damage spreads across those sites, shifting workloads cannot guarantee recovery.
In September, AWS said it could no longer restore Bahrain-only resources and data after physical damage. Damage spread across multiple availability zones, beyond what its regional design was built to withstand. In the UAE, AWS made the same determination for one of three zones; recovery continued in the other two.
Customers initially moved workloads to other regions, limiting the strike's effects, but physical damage eventually overwhelmed that protection. The episode shows moving workloads is no guarantee: other regions depend on shared networks, supplies, and host governments. Going forward, digital resilience requires more than spreading workloads; it also depends on protecting the infrastructure that connects them.
How large is the grid constraint: what share of planned global data center capacity could be delayed through 2030?
The excerpt does not state what share of planned global data-center capacity could be delayed through 2030. It therefore cannot support a numeric estimate. It identifies a basic constraint: data-center expansion needs more substations, dependable energy, physical systems, and government coordination.
Recent attacks show why infrastructure matters, but they are not evidence of a global delay percentage. After damage spread across several Bahrain availability zones, AWS said it could not restore resources hosted only there. In the UAE, AWS made the same determination for one of three zones, while recovery continued in two.
A reliable share would require forecasts or data on grid connections and planned capacity, which this excerpt does not provide. Its argument is that digital capacity is growing faster than coordination of the energy, supply-chain, and government systems supporting it. The practical takeaway is a real dependency, not a quantified 2030 shortfall: any percentage would go beyond the supplied text.
Why can moving a workload to another region fail to protect it from disruption?
Moving a workload changes its location, but does not necessarily change the systems it relies on. Regions can share transmission links, supply chains, electricity networks, and host governments. A disruption can therefore reach both the original site and its backup.
In September, AWS said it could no longer restore Bahrain resources after damage spread across multiple availability zones. In the UAE, AWS made the same determination for one of three zones while recovery continued in the other two. The attacks showed regional failover can buy time, but cannot overcome damage affecting connected facilities and infrastructure.
A faulty Google Cloud update in June 2025 disrupted services worldwide in every region. Gulf cables run through the Strait of Hormuz. In October 2026, two Iranian officials said Tehran planned to target them and impede repairs. Backups help, but real resilience requires diverse connections and infrastructure—not just another region.
Which parts of digital infrastructure depend on governments and public systems, rather than on cloud companies alone?
Cloud companies run data centers, but they cannot independently secure every system that keeps them working. Digital capacity also relies on electricity, networks, supply chains, public agencies, and host governments. Coordinating those systems crosses jurisdictions and commercial interests.
After damage spread across multiple Bahrain zones, AWS said in September it could not restore resources hosted only there. Customers could fail over elsewhere, but still depended on shared networks, supplies, and host governments beyond AWS's control.
A faulty June 2025 Google Cloud update disrupted services in every region, exposing another shared failure point. Iranian officials told Reuters in October 2026 that Tehran had plans to target Gulf undersea cables and delay repairs. Resilience therefore requires coordination among governments, utilities, network operators, and cloud providers—not just more data centers.
How do electricity and network connections allow a data center to turn computing equipment into services people can use?
Servers turn stored information and software into useful services only when they have steady electricity and working network links. Power runs the computing equipment and cooling; connections carry requests to the data center and send results back. Either failure can make a service unreachable, even if the machines themselves remain intact.
In June 2025, a faulty Google Cloud update crashed its policy-checking system across every region within seconds. It disrupted Google Cloud and Google Workspace worldwide, showing that shared digital systems can defeat geographic backup. After attacks on Gulf facilities, AWS customers failed over to other regions, but backups still relied on physical links.
Those links include subsea cables such as Asia-Africa-Europe 1, FALCON, and Gulf Bridge System through the Strait of Hormuz. Data centers therefore need dependable energy, diverse routes, and backup capacity—not merely more servers. The article warns that operators do not control every cable, supply chain, or host government, making resilience a shared infrastructure challenge.
This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.
Read more in the JupiteX app
Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.
Or read more news on the web