Chapter 2.6 — Interconnect & Networking

A modern AI cluster is not one computer but tens of thousands of chips that have to behave as one, and the wiring that binds them together has become both a bottleneck and a booming market in its own right. 650 Group forecast data-center AI-networking vendor revenue above $25B in 2028; that is a narrower and more defensible measure than the former $200B “TAM” used in this chapter. Nvidia owns the proprietary version of this plumbing; an open-standards coalition is trying to break it; and Broadcom sits in the rare position of winning almost regardless of who prevails. The next frontier is moving the optics that carry the signals onto the chip package itself.

Training a frontier model spreads the work across a cluster that can number more than a hundred thousand accelerators, and a cluster is only as fast as the slowest link between its chips. Interconnect is that link, and it comes in two layers. Scale-up is the very fast, short-range fabric that binds the dozens of GPUs inside a single rack into one giant processor; Nvidia's version is called NVLink. Scale-out is the broader network that connects thousands of racks across a data center; here the contest is between Nvidia's InfiniBand and ordinary Ethernet. As clusters grow, networking claims a rising share of the total bill, which is why this once-obscure layer is now a market that the biggest names in silicon are fighting over.

A >$25B vendor-revenue market, and a proprietary-versus-open war

The prize is large and growing fast.

The competitive structure is a familiar one: an incumbent with a proprietary, high-performance stack, and a coalition trying to pry it open. Nvidia's NVLink for scale-up and InfiniBand for scale-out are fast, mature, and locked to Nvidia's ecosystem, and they are a meaningful part of why buying Nvidia means buying a whole system. Against them, an industry coalition has built two open standards: UALink for scale-up and Ultra Ethernet for scale-out, both designed to let buyers mix hardware from different vendors. The twist is that Broadcom, rather than waiting for the UALink standard, shipped its own Ethernet-based scale-up technology first, so the near-term "open" winner has been Ethernet-flavored Broadcom silicon rather than the consortium's standard. Arista, meanwhile, is betting the scale-out network on Ethernet and calling it the eventual winner.

Interconnect Technologies and Suppliers

The layer runs from the switch silicon through the connectivity components to the optics, and Broadcom appears at almost every level.

Scale-up versus scale-out: the two battles

The standards war is really two battles with different dynamics. In scale-up, Nvidia's NVLink moves roughly 1.8 terabytes a second per GPU, and the open answer, UALink, arrived late; because UALink silicon lagged, hyperscalers wanting a 2026 product leaned on Broadcom's Scale-Up Ethernet instead, delivered through its Tomahawk switches. So the near-term winner in the "open" camp has been Broadcom, not the consortium, and UALink risks being late to its own party. In scale-out, the contest is between Nvidia's InfiniBand and Ethernet, and here the momentum is clearer: Arista's leadership is betting publicly that Ethernet is the eventual winner, and its 1.6-terabit launch in 2026 is framed as an inflection point.

This is why Broadcom is the structurally advantaged name in the whole layer. It wins on the custom ASICs of Chapter 2.5, on both scale-up and scale-out Ethernet switching here, and on the co-packaged optics below, which means it collects revenue almost regardless of which hyperscaler, which chip, or which networking standard prevails. Arista owns the scale-out Ethernet opportunity, and a cluster of connectivity specialists, Astera Labs and Credo among them, supply the retimers and cables that hold these enormous fabrics together as they scale.

Optics moves onto the package

The physical limit the whole layer is running into is that pushing electrical signals fast enough over copper burns too much power and reaches only so far. The answer is optics, and 2026 is the year it moves from pluggable modules onto the switch package itself, so-called co-packaged optics. Nvidia's Spectrum-X Photonics switches, built on optical engines made with TSMC, and Broadcom's Tomahawk-based optical switch line are both arriving this year, and the same two companies that lead switching also lead optics. The transition is gradual, and moving optics onto the accelerator package rather than the switch is still years away, with the reliability of on-package lasers the technical crux, but it opens a durable new market for the laser and optical-component makers, Coherent, Lumentum, and Fabrinet.

Economics: content growth is not the same as supplier returns

Interconnect revenue can grow faster than accelerator units because larger clusters require more switch stages, retimers, cables and optical links. That attractive content-per-system direction still does not make every supplier equally attractive. Switch silicon and network operating systems capture architecture value; optical modules and cables generally face more manufacturing competition, qualification churn and price erosion. Co-packaged optics can raise technical barriers, but it can also transfer value from a replaceable module into the switch vendor's integrated platform.

The layer also carries unusually high customer and program concentration. A merchant supplier may win a large hyperscaler design and report several years of rapid growth, yet the economics can reverse when that customer internalizes a component, changes topology or shifts the next platform to a rival. Investors should therefore track gross margin, customer concentration and content per deployed accelerator alongside headline networking revenue. The $25B forecast establishes the market opportunity, not the margin pool available to every participant.¹

America's network, China's mesh

The interconnect layer is another American stronghold, owned by Nvidia, Broadcom, and Arista. China's answer is instructive and ties back to Chapter 2.5. Because its individual Ascend chips trail Nvidia's, Huawei compensates at the level of the network, building enormous clusters: its Atlas SuperPods bind thousands of Ascend chips, over eight thousand in one system, into a single machine using Huawei's own interconnect fabric. The strategy is to reach frontier-scale computing power through system-level density and networking rather than through per-chip superiority, which makes the interconnect the crucial enabler of China's whole compute effort. It is also years behind on the software and optics that make the Western fabrics efficient.

When open fabrics become deployed systems

Watch whether open Ethernet genuinely displaces Nvidia's proprietary stack or whether Nvidia's vertical integration keeps it locked at the frontier, and in particular whether the UALink standard becomes real in deployments or whether Broadcom's head-start Ethernet keeps winning the sockets. Watch Ultra Ethernet's share against InfiniBand in newly-built training clusters, the clearest measure of the open coalition's progress. Watch the volume ramp of co-packaged optics in the second half of 2026, and the reliability of the on-package lasers. And watch whether Nvidia opens NVLink, a move that would concede the standards war while trying to keep the customers.

Who captures the networking value

Broadcom (AVGO) has the broadest public-market exposure in this layer because it captures value across custom silicon, switching, and optics under several standards outcomes. Arista (ANET) is a more concentrated expression of the shift toward Ethernet in scale-out networking. Nvidia (NVDA) still owns the integrated, proprietary stack and benefits as long as buying its systems means buying its network. The connectivity specialists, Astera Labs (ALAB) and Credo (CRDO), offer higher-beta exposure to cluster growth, while Coherent (COHR), Lumentum (LITE), and Fabrinet (FN) are exposed to the co-packaged-optics transition. These are different security profiles despite sharing the same demand driver; valuation and customer concentration determine whether strategic growth reaches shareholders.

What would loosen Nvidia's grip

The bullish case for the open-networking names weakens if Nvidia's proprietary stack proves stickier than expected and NVLink and InfiniBand hold their ground at the frontier, keeping the value inside Nvidia's system rather than letting it flow to merchant switching and Ethernet. Broadcom's position is the most robust in the layer, so the main risk to it is a broad AI-capex slowdown rather than a competitive loss. And the co-packaged-optics thesis slips if the reliability of on-package optics disappoints and the industry stays on pluggable modules longer than the 2026 timeline implies, delaying the opportunity for the optical-component suppliers.

For the platform-specific translation from switch and optical content into complete U.S. and Chinese systems, see the four supply-chain teardowns in §2.13.

Sources

Linked evidence for this chapter's figures and load-bearing claims: ¹