Equiano

Equiano is Google's privately-funded submarine cable system connecting Europe and Africa. Activated in stages between 2022 and 2023, it runs roughly 15,000 km from Sesimbra in Portugal down the entire west coast of Africa to Melkbosstrand near Cape Town, with branch landings at Lagos (Nigeria), Lomé (Togo), Swakopmund (Namibia), and the remote British Overseas Territory of Saint Helena. The cable is named after Olaudah Equiano, the 18th-century writer and abolitionist born in what is now southern Nigeria.

What makes Equiano interesting from a measurement perspective is that it represents a deliberate attempt by a single hyperscaler — Google — to bypass the traditional consortium model of submarine cable financing. Instead of waiting for telcos to negotiate capacity, Google built the cable for its own traffic, then sold the surplus to local operators. This is part of a broader pattern (Curie, Dunant, Grace Hopper, Equiano, Firmina, Topaz) where the cloud companies have become primary funders of intercontinental subsea infrastructure.

Why Equiano matters for African connectivity

Before Equiano, traffic between European data centers and most West African landing points typically detoured through older systems like SAT-3/WASC, WACS, or ACE. Many of those cables share common landing points and are aging — SAT-3 dates back to 2002. Equiano roughly tripled the available capacity on the West Africa → Europe corridor and introduced direct fiber paths to landing points that previously required transit through other countries.

For end users in countries like Nigeria, Togo, and Namibia, this matters in a way that doesn't always show up in marketing slides: it's not just about raw bandwidth, it's about route diversity. When ACE has a fault (which has happened multiple times in the past five years, sometimes affecting 13 countries simultaneously), Equiano provides an alternative path that doesn't share the same chokepoints. Resilience comes from having more than one cable.

Our measurements

GeoCables tracks Equiano through RIPE Atlas measurements launched from probes we operate in Minsk, Almaty, Tbilisi, Jerusalem, and Sevastopol. The most consistent data we have is on the Sesimbra–Melkbosstrand corridor, which is the cable's longest single segment and the one most directly comparable to "raw" Equiano performance.

Two things stand out. First, the latency is rock-solid: across 59 measurements over 30 days, we never saw a spike, never saw packet loss, never saw a routing flap. For a 15,000-km undersea path crossing the equator twice, that level of stability is the engineering goal of these systems and Equiano delivers it.

Second, there's a small but persistent asymmetry: the Cape Town → Lisbon direction is consistently 15-20 ms faster than the reverse. That's not unusual on long submarine paths — return traffic often takes a slightly different routing inside the destination country, and BGP path selection isn't symmetrical. The point is that the asymmetry is stable and predictable, which is what you want from a well-engineered backbone.

Theory vs reality: where do those 200 ms go?

The Sesimbra–Melkbosstrand great-circle distance is roughly 8,000 km. Light in optical fiber travels at about 200,000 km/s — two-thirds of vacuum speed of light, because of the refractive index of glass. The theoretical minimum round-trip time on a perfect 8,000 km path is therefore:

(8,000 km × 2) ÷ 200,000 km/s = 80 ms

Our measured average is 205 ms — about 2.5 times the theoretical floor. Where does the other 125 ms go? A few sources, in rough order of contribution:

• Cable routing isn't a straight line. Equiano follows the curve of the African coast, lands and re-amplifies at multiple points, and includes spurs that the traffic may briefly traverse. The actual fiber length is closer to 12,000 km than 8,000.
• Optical amplifiers sit roughly every 60-80 km and each adds a few microseconds. Across a transcontinental cable with hundreds of amplifiers, this adds up to a few milliseconds.
• Backhaul on both ends. The probe in Portugal isn't physically at the Sesimbra cable landing station — it's in a data center somewhere, connected via terrestrial fiber that may run tens or hundreds of kilometers before reaching the wet plant.
• Internet exchange and BGP overhead. Even on a "direct" cable, the actual IP path may make small detours through Internet exchange points where networks peer with each other.
• Protocol and host stack — ICMP processing on the target endpoint, queue depths, scheduling jitter.

A 2.5x ratio between theoretical and measured RTT is normal for transoceanic links. Our other measurements across Atlantic and Pacific cables show similar multipliers. The cable itself is rarely the bottleneck once it's built — the bottleneck is everything that wraps around it.

What we are watching

Equiano is one of the more stable cables in our monitoring set, so we don't expect frequent alerts. What we do expect to track over the coming months:

• RTT changes correlated with the activation of new spurs. Each landing brings new local routing options, which can change the IP path even when the underlying cable is the same.
• Performance from our Jerusalem and Tbilisi probes, which currently route to Africa via European hubs. As more African ISPs peer directly at IXPs in Marseille and Frankfurt, we expect those paths to shorten.
• Comparative measurements against the older African cables (ACE, WACS, SAT-3) — when one of them has an outage, we should see Equiano-routed paths stay flat while alternatives spike. That's the resilience story we want to document.

If you want to follow Equiano in real time, the live cable health monitor on our home page surfaces any RTT anomaly we catch, and the route calculator lets you see whether a given source-destination pair is likely to traverse this specific cable.

Design notes

Equiano was built with twelve fiber pairs using space-division multiplexing (SDM) — at the time of its activation it was one of the highest-fiber-count subsea cables in service. Total designed capacity is around 144 terabits per second. The cable was manufactured and laid by Alcatel Submarine Networks (now Nokia ASN), the same vendor behind most of Google's other private cables. The branching units that route traffic to Lagos, Lomé, Swakopmund, and Saint Helena are remotely reconfigurable, which gives the operator some flexibility to rebalance capacity between landings as demand grows in different markets.

One subtle but interesting design choice: Equiano's spur to Saint Helena is the first time the island has had a fiber-optic submarine cable. Before 2022, the island's roughly 4,500 residents relied entirely on satellite for international connectivity. The Saint Helena landing was funded in part by the European Development Fund as a digital-inclusion project — a reminder that submarine cables are not purely a commercial story, they are also infrastructure decisions with measurable consequences for who gets to participate in the modern internet.