CPO, Fiber Plant & Structured Cabling

By this point in Part 8 the fabric topology is decided, the switch silicon is selected, and the link budget is understood (Chapter 8.9). What remains is the most physical layer of all: the glass that carries the photons, the connectors that join it, and the increasingly fraught question of where the laser lives. An AI back-end fabric at 100k-GPU scale is a structured-cabling project measured in tens of thousands of fiber-miles and millions of mated connections — and at 800G and 1.6T per link, the optics that drive it can consume enough power and cost to justify redesigning the switch around them. Count optical ends, host conditioning and cooling together: moving the engine can shrink one line of the bill while making a failed part harder to isolate and replace.

Three choices, tightly coupled, run through this chapter. Co-packaged optics (CPO) integrates the optical engine onto the switch package — the power and signal-integrity economics that justify it, and the serviceability tax it imposes. The fiber plant sets single-mode versus multimode, the reach classes, and the channel architecture the whole ramp is trunked against. And structured cabling at scale — MPO trunking, base-N systems, polarity — lives or dies on the loss budget an install either lands inside or fails acceptance against. The CPO roadmap to 2030 is consolidated in Chapter 16.2; the install-velocity and field-execution side of cabling-at-scale lives in Chapter 7.15.

The case for CPO is built from a chain of physical facts, each of which gets worse as lane rates climb from 100G to 200G to 400G per lane. The electrical signal from a switch ASIC to a faceplate pluggable traverses package balls, PCB traces, a connector, and the module's own substrate — a path whose insertion loss grows steeply with frequency. At a higher lane mode, host-board insertion loss, return loss and crosstalk must still fit the channel budget. A retiming DSP equalizes and regenerates the signal at the cost of active power; FEC corrects errors at its specified host/module boundary and is a separate function. Shortening the electrical distance can reduce that conditioning burden.

CPO attacks the path itself. In a design that places the optical engine beside the switch ASIC on the same package substrate, the electrical run shrinks from board-scale tens of centimeters to package-scale millimeters. That shorter run can reduce the equalization needed to recover the signal; the power saving depends on which DSP or retiming functions the implementation removes. NVIDIA’s March 2025 silicon-photonics announcement claims 3.5× better power efficiency, 63× better signal integrity and 10× better network resiliency, with 4× fewer lasers, for its comparison with pluggables; Broadcom’s October 2025 Davisson announcement claims about 70% lower optical-interconnect power. Those are vendor comparison results with different boundaries, not interchangeable installed-system guarantees. At fleet scale a measured optical saving can return megawatts to compute, but count package integration, fiber attachment, host conditioning, heat removal and diagnostics together. NVIDIA stated Spectrum-X Ethernet Photonics in production on 2026-05-31, while Broadcom’s Davisson was at early-access sampling (tile below); obtain the supported FRU inventory, repair duration, installed power and delivery commitment before buying that package boundary.

The combination of radix and lane rate makes the trade most visible at heavily populated fabric tiers: 102.4 Tb/s can be presented as 64 logical 1.6T ports or 128 logical 800G ports when the selected ASIC and system support those modes. Every populated electrical attachment and optical end adds power and cable obstruction, so compare the complete switch at its required inlet condition and permitted cooling fault. Keep its thermal envelope in Chapter 8.3, channel qualification in Chapter 8.9, and dated supplier status in Chapter 16.2. CPO earns the top-of-fabric position where measured system power and demonstrated restoration justify its integration cost; future radix or module market share cannot establish that for a site.

Read the table as separate decisions about signal conditioning and service access. A qualified faceplate alternate can simplify inventory, but it is not interchangeable merely because its cage fits. LPO and LRO retain a faceplate module while changing the host’s electrical duties; either can remain the selected architecture when its complete channel passes. Near-package and co-packaged engines shorten the electrical route, but a failed internal engine may no longer be repairable with a faceplate spare and a screwdriver; optical attach and internal access become part of the system contract. There is no inevitable march down the rows: choose the combination whose demonstrated power, failure containment and replacement procedure fit the owner’s constraints.

The operational fact is simple: lasers and optical engines fail, and the failed component determines what must be drained to replace it. A faceplate module can be removed without opening the switch, but every link it carries goes with it, including a breakout assembly’s links. An internal optical-engine fault can instead require pulling a subassembly or whole board: lose the affected ports, re-cable, restart and requalify. Time the full route from alarm and diagnosis through isolation, access, replacement and loaded acceptance; insertion time is not restoration time. If the surviving fabric cannot carry admitted work through that procedure, the power saving buys a workload outage.

An External Laser Small Form-factor Pluggable (ELSFP) module keeps a serviceable light source at the front panel, feeding light over fiber to the engine that modulates and detects signals. The OIF ELSFP agreement below defines a replaceable external-laser interface; Broadcom’s Davisson record also names field-replaceable laser modules. Those facts do not show that an engine or fiber attach can be changed live. Obtain the supplier’s mapping of each laser to engines and ports, permitted isolation sequence, internal assembly FRU, board fallback and requalification procedure. For a detachable optical subassembly, verify the exact system and service manual rather than transferring the label from another platform. Spares, RMA ownership, contamination control and technician access belong in the purchase contract because the optical integration is a system dependency: you are buying a supported system, not just a replacement optic from a competitive module market.

Beneath the transceiver decision sits a more durable one: what glass you pull. Fiber outlives several generations of optics, so the fiber-plant choice is a structural decision that the density ramp must be designed against — get it wrong and the next speed step is a re-pull, not a transceiver swap. The fork is single-mode versus multimode.

Multimode (OM3/OM4/OM5) uses a wider core and short-reach VCSEL optics; modal dispersion, wavelength, fiber grade and connector loss set its application envelope. OM4 supports 100GBASE-SR4 to 100 m. In IEEE 802.3df’s 2024 800GBASE-SR8 application, nominal reach stays at 100 m on OM4 and drops to 60 m on OM3; the corresponding maximum channel insertion losses are 1.8 dB and 1.7 dB. The shorter 800GBASE-VR8 application reaches 50 m on OM4. Those are distinct application budgets, not a rule that every speed generation halves reach. Retain installed multimode when the next application closes with margin and endpoint economics favor it. Otherwise, a route that used to work becomes a new optical-application purchase or a new fiber pull.

Single-mode (OS2) uses a narrow core and laser-based transceivers; its reach is governed by attenuation and chromatic dispersion rather than multimode modal dispersion. One exact application is 800GBASE-DR8: eight parallel single-mode lanes over 16 fibers using an MPO-16 interface. A future 1.6T or 3.2T application can reuse an installed OS2 channel only when its lane and wavelength map, duplex or parallel fiber count, connector and polarity scheme, channel loss and reflectance, reach class, module, and host requirements all close on the certified plant. OS2 headroom does not guarantee a transceiver-only upgrade. Choose the long-lived fiber and topology against an explicit migration map, then certify and preserve the installed channel records.

The reach classes map cleanly onto the fabric hierarchy. Inside the rack, copper still wins where it can — passive DAC on short qualified routes, active copper where its conditioning buys the extra span (Chapter 8.9) — and the worst-case NVL72 in-rack span is short enough to keep copper viable for scale-up today. Beyond that, single-mode DR over OS2 is an intra-hall candidate, FR/LR a longer campus candidate, and coherent ZR/ZR+ an inter-campus candidate when each named application closes the installed distance, fiber, patching and repair budget; the inter-campus service belongs to scale-across (Chapter 8.8). The fiber plant is therefore a layered system: copper at the bottom for the cheapest, shortest, highest-volume links; single-mode glass where the photon’s route or the migration plan outgrows the qualified multimode application, with density and service access priced alongside reach.

A parallel-optic SR8/DR8 application carries eight optical lanes on separate transmit/receive paths, often arranged as a ribbon; a WDM FR4/LR4 application combines four wavelengths onto a duplex fiber pair. Keep the 800G or 1.6T rate and exact PMD definition attached to that lane map, including draft status where applicable. An 800G-DR8 link runs 8 transmit and 8 receive lanes — 16 fibers — terminated in a multi-fiber push-on connector (MPO/MTP). At 100k-GPU scale the fabric is a sea of these multi-fiber connectors, and the way they are grouped into trunks, the polarity scheme that keeps transmit aligned to receive, and the connector count in each channel become the load-bearing engineering of the whole physical layer.

The base-N question is which fiber granularity the trunk is built around. Base-8 (MPO-8/MPO-12 wired as 8) is an eight-fiber granularity: it divides evenly into four-pair applications and can feed a supported 400G/800G breakout without a conversion cassette when equipment cords and the actual port map match; count any stranded fibers and cassettes before claiming none. It is not a one-connector fit for an SR8/DR8 interface, which needs eight transmit and eight receive fibers — sixteen — so a DR8 port is served by two Base-8 groups whose pairing and polarity you draw explicitly on the connectivity map. Base-12, the legacy default, leaves fibers stranded or requires conversion modules when feeding 8-lane optics. MPO-16 carries 16 fibers in one connector, matching an 800G-SR8/DR8 link end-to-end in a single ferrule, and at a nominal 200G per optical lane, eight lanes give a 1.6T DR8 candidate with the same sixteen active fibers — but reusing the MPO-16 plant also requires the endpoint, connector and channel to pass that exact application. A later 3.2T step needs its own lane map and budget. The consequence of getting base-N wrong is stranded fiber, mandatory conversion cassettes (each adding a mated pair of loss), and a trunk that does not breakout to match the transceiver — re-work measured in days across a hall.

Polarity — ensuring every transmit fiber lands on the far-end receive — is the silent source of acceptance failures. The TIA-defined Method A/B/C polarity schemes each require a disciplined, documented choice of trunk type, cassette type, and patch-cord type; mix two methods in one channel and links go dark in ways that are tedious to isolate at scale. This is the strongest argument for structured cabling over point-to-point: a structured plant fixes one polarity method, documents it, and uses keyed components so a 100k-link fabric is repeatable and auditable rather than a per-link debugging exercise.

Lane-to-fiber migration record. Keep host electrical lanes, optical lanes per direction and installed fiber strands in separate fields. Record any gearbox, optical wavelength map and breakout grouping before drawing connector positions. A coded electrical signaling rate is not the nominal MAC service allocation, and neither number counts physical fibers. Attach keyed end-face drawings with pin gender and polish for each mating pair; never derive the physical pin assignment from a generic Method A/B/C name.

The composed map carries all eight Tx/Rx pairs, so the assumed Base-16 trunk passes the fiber-count and polarity design checks. Its reserved-loss requirement is about 2.6 dB from Chapter 8.9. The two-group Base-8 alternative carries the same 16 strands, but the assumed conversions change four mated pairs to six: two × 0.50 dB adds 1.0 dB, making the unrounded reserved requirement 3.640 dB, about 3.6 dB. That exceeds the current DR8 application ceiling. Keep the Base-16 route, or use equipment cords/conversion assemblies that restore the budget. Base-8 is not intrinsically disqualified: a four-pair arrangement with the same verified continuity and loss would reverse that rejection. The proposed 1.6T service remains HOLD until its distinct PMD and loaded tests pass.

The current pluggable in this assumed single-link record is one module FRU carrying one duplex link; keep its cage and extraction clearance reachable without moving adjacent live fibers, and preserve a disjoint route during replacement. If migration moves the engine inside a switch, append the laser → engine → port → board dependency map and the actual FRU procedure before accepting reuse. A shared laser feeding both survivor paths would fail the repair contract even when all 16 fibers pass. Acquire the stocked component and demonstrate drain, replacement and requalification within the owner’s outage allowance; neither an accessible ELSFP nor a compatible ferrule proves this. The DR8 application supplies the fiber boundary and the ELSFP agreement the laser interface; Chapter 7.15 owns field installation and Chapter 13.7 records acceptance.

Structured or point-to-point is the last cabling decision, and it turns on install velocity, fault isolation, and how the next density step gets executed. Point-to-point runs a jumper directly from one transceiver faceplate to another — fewer connectors (lower loss), but every move/add/change is a re-pull, fault isolation means tracing a specific jumper through a congested tray, and there is no patch field to re-patch a density change. Structured cabling lands trunks on patch panels (MDA/HDA cross-connects), with short equipment cords from panel to switch — more connectors (more loss budget consumed) in exchange for a documented, repeatable, auditable plant where a density step is a re-patch, a fault is isolated at a panel, and the install can be pre-terminated and pre-tested as factory trunks.

At AI scale the structured choice usually wins, for a reason specific to the density ramp: the trunk infrastructure is a long-lived candidate for reuse, while equipment cords, modules and host interfaces change faster. Reuse is valid only where the next application's lane/wavelength map, duplex or parallel fiber count, connector and polarity scheme, channel loss and reflectance, reach class, module and host interface all close on the installed plant. Structured cabling is how you decouple the irreversible substrate (the trunk plant) from the reversible fit-out (the optics) — the same reversible-versus-irreversible discipline that governs the slab and the power chain (Chapter 1.1). You pay the loss-budget and connector-count price up front to buy a plant you can re-patch instead of re-pull. The countervailing cost — every mated pair eats the tight 800G/1.6T budget — is exactly why the loss-budget discipline above is non-negotiable, and why hyperscale AI plants pre-terminate, factory-test, and inspect-and-clean every connection as an acceptance gate.

Choose a structured channel only when its certified lane/polarity map, retained loss reserve and accessible FRU satisfy both this application and the declared migration. A simpler jumper or a different package wins when extra patching or shared optical dependencies consume those margins; paying for nominal future-proofing then buys stranded fibers and a harder repair.