CodingBox Documentation

NIC ports, speeds and PCIe bandwidth

A NIC's front panel promises "2 × 100G"; whether the server can actually move 200 Gb/s depends on the PCIe slot behind the card, and whether a given module works depends on the cages and the controller's lane map — exactly as on a switch, but with fewer ports and a bus that is easy to under-provision. This page lists the cages and speeds found on NICs, the rules for dual-rate ports and breakout cables, the PCIe arithmetic, the OCP 3.0 form factor, and the power and cooling limits that matter for optics on cards.

Cages and speeds on NICs

CageSpeeds seen on NICsTypical controllersNotes
SFP+10G (1G on some)Intel X520/X710, Broadcom 5741x1G support varies; check the card's spec
SFP2825G, 10G (1G rare)Intel XXV710/E810-XXV, ConnectX-4 Lx/-6 Lx, Broadcom P22525G needs FEC agreement (FEC & AN)
SFP5650G, 25G, 10GConnectX-6 Lx/Dx (some)PAM4 lanes
QSFP28100G, 4 × 25G, 2 × 50G, 40G, 4 × 10GIntel E810-CQDA2, ConnectX-5/6, Broadcom P2100breakout via cable only where the controller allows
QSFP56200G, 100G, 2 × 100GConnectX-6 Dx/-7, Broadcom P2200, Pensando
QSFP112400G, 200G, 100G, 4 × 100GConnectX-7, Broadcom P1400GD100G/lane; accepts QSFP28/56
QSFP-DD400G, 2 × 200G, 4 × 100GBroadcom BCM57608
OSFP (twin-port)2 × 400G, 800GConnectX-7/-8, BlueField-3one cage, two logical ports on Nvidia twin-port designs
RJ451G / 10GBASE-T / mGigX550, X710-T, BCM57416no module, PHY on the card

Cage compatibility rules are the switch ones (Port types & cages).

Dual-rate ports and speed groups

  • Most 25G NICs run each port independently at 10G or 25G — no quad grouping — but a few dual-port controllers share a SerDes group and force the same speed on both ports.
  • 100G cards often support 40G modules; 200G/400G cards support 100G modules; going down more than one generation (1G in an SFP28, 10G in a QSFP56 via QSA) is rarely supported.
  • Speed is usually auto from the module's compliance codes; forcing (ethtool -s … speed) is needed when the module is dual-rate or unrecognised (Tools).

Breakout on NICs

CaseWorks?
100G QSFP28 NIC port → 4 × 25G to a switch (DAC splitter)only if the controller supports "4 × 25G port split" mode (many Nvidia and Broadcom cards do, most Intel do not); the card then shows 4 netdevs
Switch 100G port → 4 × 25G to four NIC SFP28 portsyes — this is the common direction; the split is configured on the switch (Breakout & MPO)
400G OSFP twin-port → 2 × 400G or 4 × 200G / 8 × 100GNvidia twin-port cards: 2 logical ports per cage; further splitting depends on firmware
2 × 50G from a QSFP28 portsome controllers

PCIe: the real port speed

PCIe generationPer lane (raw)x8 usable (≈)x16 usable (≈)
Gen 38 GT/s56–63 Gb/s112–126 Gb/s
Gen 416 GT/s112–126 Gb/s225–252 Gb/s
Gen 532 GT/s225–252 Gb/s450–504 Gb/s
Gen 664 GT/s~450–500 Gb/s~900–1 000 Gb/s

Usable throughput is 85–90 % of raw after encoding and packet overhead. Match ports to lanes:

CardPortsNeeds
2 × 25G50 Gb/sGen 3 x8
2 × 100G200 Gb/sGen 4 x16 (Gen 3 x16 saturates at ~126)
1 × 200G / 2 × 100G Gen 4200 Gb/sGen 4 x16
2 × 200G400 Gb/sGen 5 x16
1 × 400G400 Gb/sGen 5 x16 (or two x16 via "socket direct"/multi-host)
2 × 400G800 Gb/sGen 6 x16 or two Gen 5 x16

A 2 × 100G card in a Gen 3 x8 slot links at 100G on both ports and delivers ~60 Gb/s total — the optics are fine, the bus is not. Check lspci -vv for the negotiated link width and speed (LnkSta), and BIOS bifurcation settings.

OCP NIC 3.0

ItemSFF (small form factor)LFF (large)
Size76 × 115 mm139 × 115 mm
PCIeup to x16up to x32 (two x16)
Powerup to 80 Wup to 150 W
Ports2–4 × SFP28/QSFP28/QSFP56/QSFP1122 × QSFP-DD/OSFP or more
Optics accessfront of the module, tool-lesssame

OCP cards sit at the rear of the server with their own airflow channel; module cooling is usually better than on a PCIe card buried between GPUs.

Power and cooling for optics on cards

ItemLimit / rule
PCIe slot power25 W (x1…x8 low profile), 75 W (x16); 400G cards use an 8-pin auxiliary connector
Per-cage budget on NICsSFP28: 1.5 W (level II); QSFP28: 3.5–4.5 W; QSFP56/112: 5–8 W; QSFP-DD/OSFP: 12–20 W — the card's datasheet states it (Power & consumption)
Airflow200–300 LFM across the card; optics on a card in a 1U server near the CPU exhaust run hot — DDM temperature above 60 °C is common
Thermal throttlingsome controllers lower the module to low power or shut the port when the module reports high temperature
Coherent / ZR on NICsnot supported on standard NICs — class 8 exceeds the cage budget

Multi-host and socket direct

High-end cards can present themselves to two CPU sockets (two x8 or x16 links) or to several hosts. The optical side is unchanged; the split is in PCIe. Relevant for 400G on Gen 4 platforms and for GPU servers where each GPU's NIC must sit on its own PCIe root.

Checklist before buying a NIC for optics

  1. Cage type and the speeds the controller supports at that cage.
  2. PCIe generation and lanes in the target slot vs total port bandwidth.
  3. Breakout needed? Controller must support port split.
  4. Module policy of the driver (Optics compatibility on NICs).
  5. Cage power budget vs planned modules; auxiliary power connector present.
  6. Airflow in the chassis position; consider OCP 3.0 for better cooling.
  7. OS/driver support for FEC modes and speed forcing.

In CodingBox

The module's compliance codes and power class tell you whether it can serve the NIC's cage and speed; CodingBox reads them before the module is installed, and the port-split and PCIe questions are settled on the server side (Check transceiver).