Most people picture the internet as something wireless, floating invisibly through the air or bouncing off satellites in space. The reality is far more physical. The overwhelming majority of international internet traffic travels through cables laid across the bottom of the ocean. These undersea cables are the true backbone of the global internet, and almost everything you do online across borders depends on them.
What these cables are
A modern submarine cable is a bundle of hair-thin glass fibers wrapped in layers of protective material. Despite carrying an astonishing amount of data, the cable itself is often no thicker than a garden hose across most of its length, with heavier armor near the shore where the risks of damage are greatest. Data travels through the glass fibers as pulses of light, which is why the technology is called fiber optics. Light can carry enormous quantities of information and travels extraordinarily fast, which is why fiber outperforms other options for long-distance links.
Because light signals fade over very long distances, the cables include repeaters spaced along their length. These devices boost the signal so it can continue across an entire ocean without weakening into noise. Powering these repeaters across thousands of miles is one of the harder engineering challenges of building a cable.
Why not just use satellites
Satellites do carry internet traffic, and newer constellations of low-orbit satellites have improved what they can do. But undersea cables still dominate for good reasons:
- Capacity: a single modern cable can carry vastly more data than satellite links, enough for entire regions.
- Speed: light through fiber travels a short physical path compared with a signal bouncing up to orbit and back, so latency is lower.
- Cost per bit: once laid, cables move enormous volumes of data very cheaply.
For these reasons, cables handle the bulk of intercontinental traffic, while satellites are especially valuable for remote areas, ships, aircraft, and places where laying cable is impractical.
How the network is built
Laying a cable is a slow, expensive undertaking. Specialized ships spool cable off the deck and lay it along a surveyed route on the seabed, sometimes burying it in shallower waters to protect it. A single system can stretch across an entire ocean and cost a great deal to build. Because no one wants to depend on a single line, the global map is a dense mesh of many cables, so that traffic can be rerouted if one is damaged.
Ownership has shifted over time. Cables were traditionally built by consortiums of telecom companies. In recent years, large technology companies have invested heavily in their own cables to carry the traffic of their services, changing who controls this critical infrastructure.
What can go wrong
Cables break more often than most people realize. The most common causes are surprisingly mundane: fishing gear dragging along the seabed and ship anchors snagging a line. Undersea landslides and earthquakes can also cut cables. When a break happens, specialized repair ships locate the fault, haul the cable up, and splice it back together, a process that can take days or weeks depending on location and weather.
Because so much traffic depends on these lines, their security has become a strategic concern. Damage to several cables at once, whether accidental or deliberate, can slow or disrupt connectivity for whole regions. This has pushed governments to pay closer attention to who builds cables, where they land, and how they are protected.
Why it matters
The next time a video call connects smoothly to someone on another continent, it is worth remembering the physical journey behind it. The signal likely raced down a glass fiber resting on the ocean floor, through repeaters in the deep, and back up to shore thousands of miles away, all in a fraction of a second. This hidden network is one of the great pieces of infrastructure of the modern world, quietly carrying the conversations, commerce, and culture of the entire planet.