Skip to content
AI supply chain term

DWDM

DWDM (dense wavelength-division multiplexing) is a technique that sends many separate data streams down one optical fiber at once, each carried on a slightly different wavelength, or color, of light.

What it means

DWDM lets a single strand of fiber carry dozens of independent channels simultaneously by assigning each to its own tightly spaced wavelength, multiplying the capacity of installed fiber without laying new cable. It sits in the optical-interconnect layer of the AI supply chain, linking data centers to one another across metro and long-haul distances, and it is the workhorse of subsea and terrestrial backbones. As AI training and inference spread across multiple buildings, regions and sovereign clouds, the traffic between sites rides DWDM systems. It is a lever because it stretches scarce fiber and long-haul routes much further; it is a constraint because the transponders, amplifiers and precise lasers involved are specialized, and cross-region bandwidth is ultimately bounded by the physical routes that already exist.

Why it matters to investors

DWDM is how AI workloads move between sites when a single building is not enough, so demand tracks multi-region and cross-border buildouts. Systems supplier Nokia (NOK), fiber maker Corning (GLW), and optical-component makers Accelink (002281.SZ) and Eoptolink (300502.SZ) are positioned along this long-haul layer.

Companies on this part of the chain

Named to show where the term sits in the AI supply chain — research, not advice, and never a recommendation to buy or sell.

Related terms

See DWDM in the live AI chain.

THE ENTITY maps every constraint onto one live model — which part is tight now, who owns it, and who gets squeezed when it moves. Plain-English reads you can check.

THE ENTITY is an educational read on the AI supply chain — research, not investment advice. It explains how the chain works and who sits where, never price targets or buy/sell calls.