40GbE - The Older Generation, Often the Cheapest Switched Tier
Quick answer
A two-machine 40GbE link costs $679.29 today - two ConnectX-4 cards and one QSFP+ DAC, no switch. A switched four-host setup runs $1535.06 ($383.76/host) - the cheapest of the three speed tiers we track, computed the same way for each.
The used-switch economics, computed rather than asserted
The honest framing for 40GbE is not about the standard - it is a real Ethernet speed with real, documented trade-offs against 25G and 100G - it is about where it sits on the used market today. Ranking the same switched four-host build (one switch, four cards, four matching cables, all from live in-stock listings) across all three speed tiers we track:
Speed tier
Switched build, 4 hosts
40GbE
$1535.06
25GbE
$2058.92
100GbE
$3026.92
Live, not fixed - if stock shifts, this table and the ranking above it update the next time the page renders. The usual driver of the 40GbE result is switch pricing: used Arista 7050-family and Dell S4048-ON switches in our catalogue sit in the low hundreds of dollars, against thousands for the Mellanox/QNAP switches at the 100G end and roughly a thousand for the current-production QNAP switch at the 25G end.
Passive direct-attach copper, by length. A DAC is a fixed-length cable with the transceivers moulded on - you buy the whole run at once rather than a cable plus two modules.
QSFP+ transceivers exist for this tier - our table above tracks 40GBASE-SR4 multimode modules - but the selection is narrower than the SFP28 tier above it, reflecting a smaller, older, more specialised market rather than anything about the standard itself.
40GBASE-SR4 uses an MPO/MTP multi-fibre connector rather than the duplex LC connector the 10G and 25G optics in this cluster use - a physically different fibre patch lead, not just a different transceiver. Check that your fibre plant (or patch cable purchase) matches MPO/MTP before assuming an existing LC run works with a 40G optic.
As everywhere else in this cluster, a passive QSFP+ DAC sidesteps the whole optics question for a short run, and is the cheaper, cooler default recommendation for two machines near each other.
Breakout behaviour, per the standard
A QSFP+ 40G port is four 10G lanes bundled into one connector, and a breakout DAC splits them into four independent SFP+ legs - the same relationship a 100G QSFP28 port has with four 25G legs, one tier up. Live 40G-to-4x10G breakout listings are in the DAC table above and in the cabling guide.
Whether a specific switch or card actually supports breaking out a given port is a per-model, vendor-documented fact rather than something guaranteed by the connector alone - check the switch's own documentation before buying a breakout cable expecting it to work on every port.
Where to go next
The 25/40/100GbE hub has the full cluster and the form-factor map · 100GbE is the tier above this one, where breakout cabling usually is the point · ConnectX-4 is the card this page's live builds use · the cabling guide covers every DAC, AOC and optic across all three tiers · rack disk shelves are common hardware behind a build like this.
Frequently asked questions
Is 40GbE obsolete now that 25G and 100G exist?
Not obsolete, but it is the older generation of the three - QSFP+ predates SFP28's 25G-per-lane standard and QSFP28's 100G one. What that gets you on the used market is real: our own catalogue's cheapest switched build is the 40GbE one, driven by used Arista and Dell enterprise switches priced well under the 25G and 100G alternatives. See the comparison below, computed live rather than asserted.
What card do I need for a 40GbE build?
A QSFP+ (or QSFP28-cage-at-40G) dual-port card - ConnectX-4 is the one we track, covered on its own page. Coverage is thin in our catalogue (one verified listing at the time of writing), so check the live table below before assuming broad stock.
Do 40GbE optics use QSFP+ or QSFP28 transceivers?
QSFP+. The transceiver and DAC cable tables on this page are all QSFP+ parts - a 40GBASE-SR4 module needs a QSFP+ cage, and will not physically seat correctly as a 100G QSFP28 replacement despite the superficially similar name.
Can a 40G port break out to four 10G ports?
Yes, per the standard - a QSFP+ port is electrically four 10G lanes bundled into one connector, and a breakout DAC splits them back into four independent SFP+ legs. Live breakout listings are in the cabling guide and in the table below.
Are the used switches in your table still supported by their maker?
Not necessarily - check each one. Arista's own End-of-Sale advisory archive carries a 2016 notice retiring select 7050SX/7050TX/7050QX models, so a used 7050-family switch may already be off its maker's active sale list. That is a stated fact about ordering/support status, not a claim the hardware has stopped working.
Sources
Specification claims on this page were checked against the pages below on 4 August 2026. Where a widely-repeated figure did not survive that check, we report what the source says now and say so in the text rather than repeating the familiar number.
Used for: that ConnectX-3 has reached end-of-life status with NVIDIA, and the MCX311A / MCX312A / MCX314A part-number family. EOL means no new firmware or driver work from the maker; it does not mean the card stops working, and the page says so in those terms.
Used for: that mlx4 (ConnectX-3), mlx5 (ConnectX-4 and later), ixgbe (Intel 82599/X520, X540, X550) and i40e (X710) ship in the mainline Linux kernel. This is why an end-of-life card can still be a sane purchase, and it is a documented fact rather than a forum claim.
Used for: every price on this page, read from our catalogue at request time. Seeded 11 August 2026: 24 network cards across 6 models, 24 DAC cables, 8 transceivers. Every card is split tier 1 (the listing is that part) or tier 2 (a third-party board asserting that silicon).
Used for: We publish no per-port wattage for either cabling path. The relationship - 10GBASE-T draws more power and runs hotter than SFP+ direct attach - is documented at product level and is stated here. The magnitude is not, because we have not fetched and logged a source we can cite for it. Separately: every 'compatible with' phrase shown on this site is stored as a seller claim at tier 2 and badged wherever it renders. We have not tested any module against any switch.
Used for: that NVIDIA's own manuals state "The ConnectX-5 product line has moved to end-of-life" and "The ConnectX-4 Lx product line has moved to end-of-life" in those exact words (connectx5enhw and connectx4lxhw respectively, both dated 2026). The standard ConnectX-4 EN manual (connectx4enhw) carries an "EOL'd Ordering Part Numbers" table for every MCX413A/414A/415A/416A SKU but not the same plain-English banner sentence - stated as that asymmetry rather than smoothed into one claim. Also the per-SKU PCIe generation and lane count (PCIe 3.0 x8 or x16 depending on SKU; PCIe 3.0/4.0 on ConnectX-5 Ex variants) from the same manuals.
Used for: that ConnectX-4, ConnectX-4 Lx and ConnectX-5 all use the mlx5 driver family per NVIDIA's own "Supported Uplinks to Servers" table, dated June 2026 - current documentation, confirming the in-kernel driver is still maintained for hardware NVIDIA itself calls end-of-life for ordering. Also that NVIDIA's current DOCA-Host device list (docs dated September 2026) includes ConnectX-5 and ConnectX-4 Lx but not the standard ConnectX-4 - a real difference in ongoing software-package support between the two, stated as found rather than smoothed over.
Used for: that the i40e driver explicitly lists Intel Ethernet Controller XXV710 as supported hardware alongside X710/XL710/X722 (same page also notes XXV710's speed cannot be forced via ethtool - auto-negotiation only). The companion ice driver page (device_drivers/ethernet/intel/ice.html) is titled for the "800 Series" - the family E810 belongs to - without naming E810 verbatim in the rendered portion, which is stated as the limit of what this specific page confirms rather than stretched into a direct quote. Intel's own ARK product pages did not render in this sourcing pass (see the 100GbE and NIC pages' own note); PCIe generation and EOL status for XXV710 and E810 parts are marked not sourced for that reason.
Used for: the 7050X family's own published figures: "10GbE Typical Power: 1.2W per port", and for the 7050QX-32S specifically "Typical Power Draw: 150W (4.5W per port)" - the only per-port power figures this cluster publishes, because they are the only ones a vendor datasheet actually rendered and stated. Also Arista's own End-of-Sale advisory archive (arista.com/en/support/advisories-notices/endofsale), which carries a dated 2016 end-of-sale notice for "Select Models of the Arista 7050SX, 7050TX and 7050QX Series" and, separately, 2026-dated end-of-sale notices for the newer 7050SX3-48YC12 and 7050SX3-48YC8/48C8 models - a used 7050-family switch may be running silicon its own maker has since end-of-saled, stated with the notice's own date rather than ours.
Used for: every price in this cluster's tables, read from our catalogue at request time. Seeded 13 September 2026: 12 NICs across 3 models (ConnectX-5, ConnectX-4, Intel E810), 40 DAC cables (25G/40G/100G including breakout), 18 transceivers, 11 used-enterprise and current-SMB switches. Two vendor documentation gaps stated plainly rather than filled: NVIDIA/Mellanox's per-model switch datasheets (SN2010/SN2100/SN2410/SN2700) did not render in this sourcing pass beyond a series-level marketing page, so those four switches carry no sourced power, acoustic or EOL figures; Intel's ark.intel.com did not render at all in this pass, so XXV710 and E810 PCIe generation and formal EOL status are marked not sourced rather than assumed from the 700/800-series family pattern.