Petal puts multi-core fibre on the transatlantic timetable

On 21 September, Meta announced Petal, a planned 7,000-kilometre submarine system between the United States and France that is expected to enter service in 2029. The project targets 1 petabit per second, or 1,000 Tbps, using two-core fibre in a 24-fibre-pair cable. NEC will design and build the system, Sumitomo Electric will supply the fibre, and Orange is supporting the French landing and terrestrial interconnection. The headline is large, but the more consequential change is architectural: multi-core fibre is moving from a laboratory proposition into a commercial transoceanic build.

That matters to carriers far beyond the route itself. Petal will not create a new landing point in the Balkans. It will, however, test whether the next doubling of submarine capacity comes from more fibre pairs, wider optical bands, or more spatial channels inside the same physical cable. Those choices will eventually determine what European carriers can buy, protect and route toward Southeast Europe.

The number is 1 Pbps. The design is the story.

Meta describes Petal as the first transoceanic cable designed to deliver petabit capacity and says it will carry twice the capacity of today’s most advanced systems at this distance (Meta Engineering). Those are project claims, not a measured service result: the cable is still planned, and the announced target is 2029.

The capacity target comes from combining 24 fibre pairs with two cores in each fibre. In a conventional single-core arrangement, counter-propagating traffic uses a pair of fibres. Petal’s two-core fibre places two optical paths in one fibre strand. Meta describes the resulting system as equivalent to 48 fibre pairs, although that does not make it interchangeable with a conventional 48-pair cable for every operational or commercial purpose.

That distinction is important. A design-capacity number is not the same thing as lit capacity, saleable spectrum or an end-to-end service protected across multiple networks. Petal’s announcement does not state how much capacity will be offered to third parties, whether any spectrum will be sold, or what commercial interfaces will be available. The project is owned and operated by Meta, rather than announced as an open wholesale system (NEC).

A spatial-division bet below the waterline

The engineering challenge is to add spatial channels without turning the cable into a larger, less efficient system. Meta says the two-core fibre keeps the outer diameter at 125 micrometres, while Sumitomo’s fibre is designed to keep attenuation low and crosstalk between the cores practically immeasurable (Meta Engineering). Those properties are necessary for the system to behave like a deployable long-haul platform rather than a short laboratory link.

The repeater is just as important as the fibre. A 7,000-kilometre cable normally needs roughly 100 repeaters. Petal’s design uses a single-body repeater with 96 amplifiers: 24 fibre pairs multiplied by two cores and two directions. A fan-in/fan-out interface converts the two-core signal into single-core paths for amplification, then converts it back. The announced design also keeps the system within existing 18 kV power-feeding equipment limits, avoiding a wider requalification of the subsea ecosystem (Meta Engineering).

This is where the claim should be tested over time. The announcements provide no full power budget, repeater spacing, repair performance data or operating history. They do show the trade-off the industry is pursuing: more capacity per cable without simply adding another complete cable, another set of repeaters and another power burden.

The owner, the builder and the landing party

Petal is also a statement about who is willing to carry the integration risk. NEC says Meta will fund and operate the cable, while NEC takes responsibility for overall design and construction and Sumitomo provides the two-core fibre (NEC). Sumitomo’s own announcement describes a 48-core-pair configuration and high-density repeater mounting with 48 amplifier pairs, with power supplied from one shore-end facility (Sumitomo Electric).

Orange’s role is not a minor landing detail. Meta says the French operator is working with it on the Atlantic landing and the terrestrial interconnection into Europe, including onward connectivity to other European countries (Meta). The parties have not named the French cable landing station, published a construction schedule more precise than 2029, or disclosed ownership and capacity-allocation terms.

That leaves a familiar gap between a highly specified wet plant and an underspecified service. The cable can be technically successful and still have limited value to a regional carrier if the European handoff is congested, if access is restricted to Meta’s own network, or if the route lacks independent terrestrial protection. Petabit capacity at sea does not remove the need for diverse backhaul, neutral interconnection and clear failure domains on land.

What it means for the region

For Bulgaria and the Balkans, Petal is an upstream capacity event, not a new local landing. Its relevance will be decided in the European network behind the French coast. If Orange and its partners create genuinely diverse paths into the continent, the system can add another high-capacity input to Western Europe. If the capacity remains primarily private, the effect will be indirect: more traffic will be absorbed by Meta’s own infrastructure while wholesale carriers compete for the remaining routes and spectrum.

Regional buyers should therefore ask different questions than “how many terabits does Petal carry?” They should ask whether capacity can be handed off without a single Paris or Marseille bottleneck; whether a transatlantic service can be paired with an independent route through Italy, the Adriatic, Greece or the Black Sea; and whether restoration is independent at the cable, landing-station and terrestrial levels. They should also separate a protected wavelength from a spectrum block or a full fibre-pair commitment. Those products have different failure modes and different capital requirements.

For Sofia Connect’s market, the practical consequence is a higher bar for route design. New upstream capacity is useful only when it can be combined with predictable IP transit, DWDM handoffs and physical diversity into Sofia and neighbouring capitals. Petal’s 2029 target is still several planning cycles away. The decision that matters now is whether Southeast European networks prepare to consume the next generation of capacity as a cheap bulk input, or as a scarce, privately controlled asset whose value depends on access terms.

Sources

• Meta: Announcing Petal, a First-of-its-Kind Transoceanic Subsea Cable
• Meta Engineering: Inside Petal: Building the World's First Petabit-Class Transoceanic Subsea Cable
• NEC: Meta, NEC, and Sumitomo Electric to Collaborate on “Petal”
• Sumitomo Electric: Petal commercial petabit-class optical submarine cable system