West Africa Cable System (WACS)

The West Africa Cable System (WACS) is a 14,530 km submarine cable constructed by Alcatel-Lucent and ready for service in 2012. It runs from Yzerfontein in South Africa's Western Cape up the entire west coast of Africa to London, with 14 landing points — 12 along the African seaboard (Namibia, Angola, both Congos, Cameroon, Nigeria, Ghana, Côte d'Ivoire, Togo, Cape Verde) plus two in Europe (Canary Islands, Portugal) before terminating on land at a cable station in London. At its commissioning, WACS cost $650 million and was designed with four fibre pairs delivering 5.12 Tbps — the first high-capacity cable connecting West and Southern Africa to Europe along a single continuous fibre body.

WACS, ACE (also 2012), and SAT-3/WASC (older, 2002) together form the West African Atlantic cable spine. Where EASSy serves the Indian Ocean coast of Africa, WACS owns the Atlantic side. If you trace the coast from Cape Town to the English Channel, WACS is the cable that touches it.

The asymmetry that points at SACS

Our monitor measures WACS between its two end landings: Yzerfontein in South Africa and Seixal in Portugal. Over 30 days we collected 48 samples across both directions, and the data has a curious property.

The forward direction (Seixal → Yzerfontein) measures 210.6 ms minimum. Light in submarine fibre has a theoretical round-trip minimum of 142.0 ms for WACS's 14,530 km length. 210.6 ms is 1.48× the physics floor — typical for a multi-landing coastal cable that spends time at each station.

The reverse direction, Yzerfontein → Seixal, measures 128.8 ms at its minimum. That is below the physics floor for a full round-trip through WACS (142 ms). Physically impossible if the packet traverses WACS end-to-end. The reverse path is using a different route.

128.8 ms corresponds to a fibre traversal of about 13,150 km. The great-circle distance from Yzerfontein to Seixal is roughly 7,600 km. Something between. A plausible reconstruction: the reverse path uses the SACS cable from South Africa's region (or via WACS's southern segment to Angola, then through SACS) across the South Atlantic to Fortaleza, then north through one of the Brazil-to-Europe cables (ellaLink, EulaLink, or similar) to Portugal. Several cables, multiple handoffs, but collectively shorter fibre path than the full WACS run up the African coast.

Nine to thirteen IP hops on the reverse direction support this: mixed-cable paths accumulate more handoff points than a single-cable traversal. And the hop count variance (9, 11, 12, 13) on different days suggests the return path selection varies day-to-day, depending on which cable has capacity and which peering is currently preferred.

14 landings, one continental backbone

Fourteen landings is unusually generous for a cable of WACS's vintage. Most 2010-era cables went for five to eight landings. WACS's 14 reflect its design philosophy: a single continuous cable that would serve the entire West African seaboard, giving each coastal country direct fibre connectivity to European exchanges without needing to transit through regional hubs.

Before WACS, several West African countries reached Europe only through SAT-3/WASC (a 2002 cable with fewer landings and lower capacity) or via satellite. WACS's 2012 launch collapsed that dependency. Overnight, countries like Cameroon, Togo, and the Congos had direct multi-gigabit-per-second connectivity to European internet exchanges — capacity that fundamentally reshaped the affordability of consumer internet across the region.

The London termination is notable. Unlike most submarine cables, which terminate at coastal landing stations, WACS's European trunk continues via terrestrial fibre to a cable station in London — giving it direct access to the UK's massive peering infrastructure without intermediate European handoffs.

Four fibre pairs and 13 years in service

WACS has four fibre pairs. In 2012, that was a solid spec — contemporary cables like SEA-ME-WE 4 had three. The original 5.12 Tbps design capacity has been upgraded multiple times through electronics refreshes, now carrying significantly more than its commissioning spec. Like most cables of its generation, the submarine body is unchanged; the transponders at each landing have stepped through successive generations of coherent modulation to increase usable capacity.

WACS's shareholders form a consortium of African, European, and global operators: MTN, Vodacom, Telkom South Africa, Cable & Wireless, Tata Communications, Angola Cables, Orange, Portugal Telecom, and several others. This distribution of ownership across regional carriers, rather than concentration in one hyperscaler or one national telecom, is characteristic of 2000s-era consortium cables. Each owner takes dedicated capacity on specific fibre pair allocations; the shared resources (power feed, landing stations, repair arrangements) operate via a joint agreement.

What WACS made possible

A number of things that seem routine in 2026 were not before 2012. West African enterprises can host services on European cloud infrastructure with single-digit-millisecond overhead compared to native European use. African ISPs can buy European transit wholesale rather than leasing satellite links. African users can stream video from European or American origins at competitive latency. None of these requires WACS specifically — multiple cables serve the region now — but WACS was the first to make any of them widely available.

The Yzerfontein landing deserves specific mention. Yzerfontein is a small town on South Africa's Western Cape, about 80 km north of Cape Town. The area has become one of Africa's densest cable-landing clusters: WACS, Equiano, 2Africa, SEACOM, and SAT-3 all land in the immediate vicinity. A 2024 cable fault on any one of them remains a regional inconvenience rather than a disaster because of the redundancy these co-landings provide.

Comparing to Equiano (a decade later)

Ten years after WACS, Google commissioned Equiano along a very similar route — Portugal to South Africa, with African coastal landings. Equiano is shorter (6,100 km, single trunk without the western-African coastal detours), higher-capacity (144 Tbps vs WACS's 5.12), and fibre-pair denser (12 pairs vs 4). Our measurements on Equiano show 2.5× physics-floor performance; WACS shows 1.48× forward and sub-floor reverse due to multi-cable routing.

Both cables serve West African connectivity. Equiano is optimised for hyperscaler data-centre traffic (Google's own services). WACS is optimised for regional telecom capacity (consumer ISPs, enterprise connectivity). The two are complementary rather than competitive — most West African operators peer on both, routing different traffic through whichever cable offers better commercial terms for that flow.

What our data proves

• WACS forward-direction (Seixal → Yzerfontein) measures at 210.6 ms, 1.48× the physics floor. Typical for a multi-landing coastal cable; consistent with the 14-station architecture.
• Reverse direction at 128.8 ms is below WACS's full-length physics floor. The return path uses a shorter combination of cables — most plausibly SACS (which we measured at near-physics-floor on the Africa-Brazil segment) plus a trans-Atlantic European return.
• Thirteen years after commissioning, WACS is still doing what it was built to do. Consistent forward-direction latency, no fault-related disruptions in our 30-day measurement window.

WACS is one of the quiet workhorses of African submarine connectivity. It is not the newest or the fastest or the widest-capacity cable, but it is the one that in 2012 made modern African internet economically possible. That we can read its architectural fingerprint directly off our ping measurements — and see its asymmetry pointing at SACS as the preferred return — is a small illustration of how interconnected the modern African cable system has become.

Try it yourself

Live data on the WACS cable page. For context on African cables, see Equiano (2022, Google, same route), SACS (2018, Africa-Brazil), and EASSy (East African coast).