Summary

  • Arelion says its Nokia-assisted field trial added about 21% to usable optical spectrum on a 500 km Amsterdam–London route.
  • The announcement does not quantify how many additional wavelengths became dependable customer capacity, what they cost to operate, or whether customer traffic used them.

The useful question after a capacity announcement is what a buyer can order. On 8 October, Arelion said it had completed a live-network field trial on its Amsterdam–London route using Nokia's 1830 GX Super C and Super L open line system, CHMQ6 thin transponder modules, and ICE-X 400G and 800G ZR+ coherent pluggable optics. Arelion reported an approximately 21% expansion in usable optical spectrum over existing fibre.

That is a meaningful engineering result. It is not yet a 21% increase in traffic, committed bandwidth, revenue, or customer-facing capacity. A fibre system's line capacity depends on more than the width of the spectrum available: it also depends on how many wavelengths fit, the line rate and spectral efficiency of each, and how the path performs. The ITU's technical report on international optical networks lays out that distinction. ITU-T G.694.1 defines the fixed and flexible DWDM frequency-grid framework in which channels and slots are arranged.

Arelion's announcement says the additional capacity can support high-bandwidth offers such as its Managed Optical Fiber Network, which links data centres and other digital infrastructure. It also describes the company as evaluating a more scalable and energy-efficient approach. The release gives no before-and-after aggregate capacity in terabits per second, channel count or mix, error and availability results, measured watts per delivered bit, or commercial turn-up date. "Live-network field trial" identifies the setting; it does not establish that customer traffic traversed the newly available spectrum.

Equipment claims need their own labels. Nokia says its 1830 GX can provide more than 25% additional fibre capacity against traditional extended C-band systems, while its CHMQ6 and ICE-X materials describe 400G and 800G modes, client splitting and lower power per bit. These are Nokia product claims and datasheet specifications, not Arelion measurements from this route. The distinction matters when an operator turns an engineering demonstration into a customer promise.

A SDxCentral report recaps the 21% figure and equipment, but publishes no route-level measurements of its own.

The commercial test is a service customers can buy and depend on. Arelion would need to show how much additional protected capacity it can provision, what latency and availability commitments apply, and whether the cost per delivered bit falls after equipment, power, operations and any fibre constraints are counted. If wavelengths can be activated only on some segments, the end-to-end bottleneck may erase the headline gain. If capacity is available but uncontracted, it remains an option rather than revenue.

The trial therefore opens a useful investment question rather than closing one. Broader spectrum can defer new fibre construction and give data-centre links more room. But the business value appears only when that room survives the route, meets an agreed service profile and is priced against customer demand. The next evidence should connect the optical result to a product sheet, a tested service path and a measurable unit cost.