The Roads

Networking & Interconnect

Layer 04 of seven · Networking & Interconnect

Why it matters
Idle accelerators are the most expensive thing in a data centre. Interconnect determines how much of the compute a buyer has paid for actually gets used, which makes it a first order economic question rather than a technical footnote.
The bottleneck
Optical component supply and the physics of moving high data rates over distance. Copper stops working sooner than people expect, and optics is where much of the incremental spend lands.
Who captures value?
Suppliers whose products are designed into a platform, and those holding hard engineering in optics and high speed signalling. Standardisation tends to erode pricing over time.
What could change?
Ethernet based fabrics maturing into a full alternative to proprietary interconnect would widen the supplier list and change pricing across the layer. Co-packaged optics, which places the optical engine next to the switch silicon, would change the supply chain again.

Training a large model means tens of thousands of processors behaving as one machine. Networking can become the bottleneck even when there is plenty of compute available.

A cluster is only as fast as its slowest exchange. During training, accelerators must repeatedly share intermediate results, and every one of them waits for the last to arrive. Add enough machines and the network, not the silicon, sets the pace. The engineering runs across four distances: GPU to GPU inside a server, server to server inside a rack, rack to rack across a hall or campus, and data centre to data centre between regions. Each distance has its own technology and its own suppliers.

GPUSSwitchARISTASwitches shuttle data between GPUs inside the building. If the switch is slow, thousands of expensive chips sit waiting.OpticsFIBRE TRANSCEIVERSTransceivers turn electrical signals into light. Coherent and Lumentum make the components that push data down the fibre.Second data centreLarge training runs increasingly span multiple buildings, so the link between sites has to behave almost like an internal cable.Thousands of GPUs must behave as one machine.

Tap or hover a box to see what it does

Inside a rack, switches move data between GPUs. Between buildings, light does the work through fibre optics.

GPU to GPU inside a server

The shortest and fastest hop. NVIDIA's NVLink and NVSwitch connect accelerators inside a server so that several GPUs behave as one larger device, with far more bandwidth than a standard network link could carry.

publicNVDA

NVIDIA

What it does

Designs the GPUs and accelerated computing platforms used for much of modern AI training and inference, and sells the networking that ties them together.

Why it matters

The hardware is only half of it. CUDA, the software libraries built on top of it, and a very large developer base make NVIDIA systems the default choice for teams who want to ship rather than port.

What could go wrong

Custom accelerators built by its own largest customers, a stronger AMD, revenue concentrated in a handful of buyers, export restrictions, and model architectures that shift the balance of demand.

Position in the ecosystem

Server to server and rack to rack

Across a rack and a hall, clusters run on InfiniBand or high speed Ethernet. NVIDIA supplies InfiniBand and its Spectrum-X Ethernet platform following the Mellanox acquisition, while Arista and Broadcom drive the open Ethernet alternative and Credo supplies the high speed connectivity between them.

publicNVDA

NVIDIA

What it does

Designs the GPUs and accelerated computing platforms used for much of modern AI training and inference, and sells the networking that ties them together.

Why it matters

The hardware is only half of it. CUDA, the software libraries built on top of it, and a very large developer base make NVIDIA systems the default choice for teams who want to ship rather than port.

What could go wrong

Custom accelerators built by its own largest customers, a stronger AMD, revenue concentrated in a handful of buyers, export restrictions, and model architectures that shift the balance of demand.

Position in the ecosystem
publicANET

Arista Networks

What it does

Builds high speed Ethernet switching and the network operating system used in large data centres and AI clusters.

Why it matters

Ethernet is the incumbent everywhere else in computing, and the industry is working hard to make it a first class option for AI clusters. Arista is the clearest listed way to hold that view.

What could go wrong

A few very large customers drive much of the revenue, and merchant switching silicon lets others compete on price.

Position in the ecosystem
publicAVGO

Broadcom

What it does

Co-designs custom AI accelerators for hyperscalers and supplies much of the switching and connectivity silicon inside data centres.

Why it matters

It sits on both sides of the custom silicon story. When a cloud provider builds its own chip to reduce dependence on merchant GPUs, Broadcom is often the partner making that possible.

What could go wrong

Custom programmes are concentrated in a few customers who can change direction, and the software business carries its own integration risk.

Position in the ecosystem
publicCRDO

Credo Technology

What it does

Makes high speed connectivity products including active electrical cables and retimers used inside racks.

Why it matters

As data rates climb, plain copper stops behaving. Active cabling is a small line item that quietly determines whether a rack works at the speed it was sold at.

What could go wrong

Small, concentrated and exposed to design cycle timing at a few customers.

Position in the ecosystem

Data centre to data centre

Beyond the hall, traffic moves over optics. Transceivers, lasers and photonic components carry campus and long haul links, and co-packaged optics is the direction of travel as data rates climb.

publicCOHR

Coherent

What it does

Makes optical transceivers, lasers and photonic components used to move data between racks and buildings.

Why it matters

Optics is where a very large share of the incremental networking spend in AI clusters lands, because copper runs out of reach quickly at these speeds.

What could go wrong

Historically a competitive, price eroding market with rapid generational transitions.

Position in the ecosystem
publicLITE

Lumentum

What it does

Supplies photonics including lasers and transceivers for data centre interconnect.

Why it matters

A second established supplier in a market where hyperscalers actively want more than one source.

What could go wrong

Legacy telecom exposure and the same pricing dynamics that affect the whole optics sector.

Position in the ecosystem
publicMRVL

Marvell Technology

What it does

Designs data infrastructure silicon including optical interconnect, custom compute and storage controllers.

Why it matters

Moving a signal off a chip, across a rack and between buildings is its own engineering discipline. Marvell sells into the parts of that path most people never think about.

What could go wrong

Custom programme timing is lumpy and dependent on a handful of hyperscaler decisions.

Position in the ecosystem
publicLWLG

Lightwave Logic

What it does

A pre-revenue research company developing polymer based electro-optic materials for high speed optical modulation.

Why it matters

If the materials work at production scale they could improve the speed and power profile of optical links. That remains a research outcome rather than a shipping product.

What could go wrong

No meaningful revenue, ongoing funding needs, and a long path from demonstration to qualified supply.

Position in the ecosystem

Region to region and out to users

Between regions, and then to the people and devices actually making requests. Latency at this distance shapes where inference is served from rather than where models are trained.

publicNET

Cloudflare

What it does

Operates a global network providing content delivery, DDoS protection, zero trust services and edge compute.

Why it matters

It sits in front of a large share of web traffic, which gives it both a security vantage point and a natural place to put inference close to users.

What could go wrong

Monetising a wide free tier is a long game, and the developer platform competes with the hyperscalers.

Position in the ecosystem

How to think about the economics

  • GrowthVery high

    How quickly demand in this part of the ecosystem is expanding.

  • Capital intensityModerate

    How much money has to be spent up front before revenue arrives.

  • Competitive moatModerate

    How difficult it is for a credible new entrant to take the business.

  • Customer concentrationVery high

    How much revenue depends on a small number of buyers.

  • Disruption riskModerate

    How exposed the layer is to a technical or commercial shift.

This is a framework for thinking about the economics of a layer, not a recommendation. An important AI company, a strategically advantaged company, an investable security and an attractively valued security are four different things.

Where value may accrue

Design wins, switching costs inside a platform, and hard physics. Once a fabric is chosen for a cluster generation it rarely changes mid-flight, which gives incumbents a useful few years.

Key terms in this layer

Fibre optics
Glass threads that carry data as pulses of light. Inside AI data centres and between them, fibre is what keeps thousands of GPUs in step.
ExampleCoherent and Lumentum make the transceivers that turn electricity into light.
Related termsNetwork switchData centre
Network switch
The hardware that shuttles data between machines. If the switch is slow, thousands of expensive GPUs sit idle waiting for data.
ExampleArista builds the switches inside many AI data centres.
Related termsFibre opticsGPUData centre
InfiniBand
A very low latency networking standard used to link accelerators inside AI clusters, supplied largely by NVIDIA since it acquired Mellanox. High speed Ethernet is the competing approach.
ExampleLarge training clusters connect thousands of GPUs over InfiniBand.
Related termsNetwork switchFibre opticsGPU

Related questions

Doesn't NVIDIA manufacture its own chips?

No, and this surprises a lot of people. NVIDIA designs the chips and outsources manufacturing, mostly to TSMC in Taiwan. It is a fabless company: engineers and intellectual property rather than factories. AMD and Apple work the same way. Physical manufacturing is extraordinarily capital intensive, and TSMC has spent decades becoming very difficult to replace at the leading edge.

What is the difference between a CPU and a GPU?

CPUs are optimised for flexible, general purpose and low latency computing. They handle branching, unpredictable work and the everyday business of running an operating system. GPUs contain many more, simpler processing units and are especially effective at performing large numbers of similar calculations in parallel. That parallel arithmetic is most of what training and inference consist of, which is why a design originally refined for rendering graphics turned out to suit AI so well.

Sources and further reading (3)+
  1. 1InfiniBand and Spectrum-X networking documentation · NVIDIA
  2. 2Investor relations and product documentation · Arista Networks
  3. 3Ultra Ethernet Consortium specification work · Ultra Ethernet Consortium
Last fact-checked: 29 August 2026

The AI ecosystem changes rapidly. Company positions, technologies and market data reflect information available at the date above.