Separate the cable from the service
International communications rely on several layers working together. Submarine cables carry a large share of intercontinental data, but an end-to-end service also depends on landing stations, terrestrial fibre, routers, power, data centres, network operations and the commercial arrangements connecting providers. A fault in one layer may affect service differently from a fault in another. Understanding resilience starts with asking which component is disrupted and which paths remain available.
A submarine cable is a physical link between points. A network operator can route traffic through different links if those links connect appropriate places and have usable capacity. That is similar to a road network: an alternate road is helpful only if it reaches the destination, remains open and can accommodate the traffic. In communications, routes must also work with addressing, routing policy, capacity planning and the systems that hand traffic between networks.
Capacity is not simply a question of whether another cable exists. If the remaining routes are heavily used, they may not carry all disrupted traffic at the normal performance level. Traffic may be rerouted, prioritised or experience congestion. Network operators plan and monitor these conditions, but no design can make every failure invisible or guarantee identical service during every incident.
Route diversity means real separation
Geographic diversity can reduce shared exposure. Two systems landing in the same coastal area may still share a vulnerable approach, landing facility or terrestrial corridor. Routes that appear separate at continental scale may converge at a narrow maritime passage. A useful resilience assessment looks at the full path: undersea corridor, shore approach, landing station, terrestrial backhaul and onward interconnection.
There are also practical dependencies that a simple map cannot show. Multiple cables may share power, equipment, buildings, maintenance arrangements or commercial bottlenecks. Operators examine these dependencies as they design networks and restoration plans. Diversity is strongest when alternate routes fail for different reasons rather than being exposed to one common event.
The geography of the North Atlantic, the English Channel, the Mediterranean and the Strait of Malacca offers different examples of distance, concentration and maritime activity. The point is not that a particular sea is inherently unsafe. Rather, a corridor’s characteristics affect the number and type of routes, landing choices and operational plans that can be made.
Redundancy needs capacity and action
Redundancy describes spare or alternate capability, but it is useful only when it can be activated and has enough capacity for the circumstances. Network operators establish routing policies, measure performance and maintain agreements that allow traffic to move between networks. A path might be physically intact yet unavailable for a particular service because it lacks the necessary capacity, connection or commercial arrangements.
Operators can prepare for disruptions through monitoring, route diversity, spare capacity, contingency planning and communication procedures. A resilience plan identifies which links can carry traffic, what degradation is acceptable, who coordinates a response and how information reaches customers and partners. A plan should be tested against realistic scenarios, including simultaneous faults or disruptions to terrestrial segments.
Redundancy has costs. Building another cable system or extending terrestrial connectivity requires capital, permits, landing infrastructure, ongoing maintenance and power. The appropriate level of resilience depends on the service, geography and consequences of interruption. Not every service needs the same protection, but the choices should be explicit rather than assumed from the appearance of a map.
A second cable is not automatically a second route; resilience lives in the independent paths and capacity behind it.
Physical repair is part of resilience
Alternate routing can keep services operating while a cable fault is addressed, but the underlying system still needs restoration. Repair agreements, access to specialised vessels, spare cable, jointing capability and reliable route records all influence recovery time. These resources are shared across a global industry, and a repair ship may be required in a different region when another fault occurs.
Repair planning therefore complements redundancy. An operator with multiple routes may still need to repair damaged infrastructure to restore reserve capacity and reduce the risk of a later incident. Conversely, efficient repair capability can limit how long a route remains unavailable. Both network-level traffic management and marine engineering are necessary parts of continuity.
The International Cable Protection Committee promotes cable protection and cooperation, and the International Telecommunication Union develops relevant technical recommendations. The United Nations Convention on the Law of the Sea provides an international framework that includes provisions related to submarine cables. Their roles differ: technical recommendations, industry practice and international legal frameworks are not interchangeable, but all help readers understand the environment in which these systems operate.
Measure what resilience means
Good questions are specific. Which landing points remain reachable? How much traffic can alternate paths accept? Do those paths share a landing station or terrestrial segment? How quickly can traffic be switched? What is the expected repair mobilisation time? Which customers or services are most sensitive to latency, packet loss or interruption? Answers reveal whether resilience is operational, merely geographic or only an assumption.
Resilience also has a human dimension. Operators coordinate with landing facilities, network partners, government agencies, maritime authorities and repair contractors. Clear incident communication can reduce uncertainty while technical teams diagnose and restore service. Documentation needs to be accurate and protected appropriately, since route and facility details have operational value.
The internet is not a single machine that can be switched off or saved by one intervention. It is a federation of networks with different paths and responsibilities. Undersea cables are a central part of that physical fabric, yet their reliability depends on the larger system around them. Continue reading Questions Across the Ocean, or return to the cable chronicle to follow the story from shore.