2.3 GHz / 16 Cores / 32 Threads / 22 MB Cache / 110 W / SST-BF
| Socket | LGA3647 |
|---|---|
| Series | Xeon Gold |
| Generation | 2nd Gen (Cascade Lake) |
| Cores | 16 |
| Memory Type | DDR4 RDIMM · DDR4 LRDIMM · DDR4 3DS RDIMM |
| Max Memory Speed | 2667 MHz |
| PCIe Version | PCIe 3.0 |
| Lithography | 14 nm |
| Memory Channels | 6 channels |
| Condition | New · Used |
| Threads | 32 |
| Base Frequency | 2.3 GHz |
| Max Turbo Frequency | 3.7 GHz |
| Cache | 22 MB |
| TDP | 110 W |
Intel Xeon Gold 5218N — 16 cores and 32 threads at 2.3 GHz base / 3.7 GHz turbo, 22 MB cache, 110 W TDP, four-socket scalability, six-channel DDR4-2667 and Speed Select Base Frequency support, in FCLGA3647 packaging. Supplied new or as tested pulls, with a 1-year warranty.
The N designation is not a speed grading, it is a deployment qualification. This is the same sixteen cores, the same 22 MB and the same 2.3 GHz base clock as the standard 5218, but the part is binned for network and embedded shelves: 110 W instead of 125 W, a 3.7 GHz turbo ceiling instead of 3.9, a case limit of 81 °C, plus two things the plain 5218 does not have — Embedded Options Available: Yes and SST–BF support. ARK still lists this SKU as Launched, where most of the 5000-series Golds are marked Discontinued.
What it is not is faster. Turbo is 200 MHz below the 5218, so lightly-threaded spikes are marginally slower, and the case limit is lower in absolute terms. Everything else is identical: 1.375 MB of cache per core, six channels at 2667, two UPI links, 48 PCIe 3.0 lanes, four-socket scalability and one AVX-512 FMA unit per core.
Run the thermal arithmetic and the point of the bin becomes clear. 110 W against an 81 °C case limit leaves 0.418 °C/W of heatsink allowance from a 35 °C inlet; the 125 W 5218 at 87 °C needs 0.416. The cooling requirement is the same to two decimal places — so the N does not buy you a smaller heatsink, it buys you 15 W back in the power budget, which is what actually limits density in a 1U node, a power-capped rack, an edge cabinet on a PoE-fed supply or a NEBS-compliant shelf counting every watt of heat rejection.
Speed Select Base Frequency is the other differentiator, and this is the only 5218 variant that carries it. SST–BF lets the platform pin a guaranteed base clock on a chosen subset of cores while running the remainder lower — the classic pattern for packet processing and control-plane work, where a few cores need deterministic latency and the rest are throughput filler. Intel's ARK entry confirms the capability but does not publish the per-profile frequencies or power for this SKU; those come from the platform firmware, so confirm the profile table with your server vendor before you design around a specific number.
AVX-512 is present with one FMA unit per core — the Gold 5000-series rule, with 5122 and 5222 the documented exceptions at two. Virtualisation, web and application tiers, databases, file serving and network functions never notice. Dense linear algebra, HPC and inference kernels compiled around 512-bit vectors do, and a Xeon Gold 6240 with its second FMA unit and 2933 memory is the correct purchase instead.
| Product collection | 2nd Gen Intel Xeon Scalable (Cascade Lake) |
|---|---|
| Processor number / sSpec | 5218N / SRGLN (L1 stepping) — an earlier SRFD9 build of the same part exists, also L1 |
| Ordering code | CD8069504384601 (tray, SRGLN) / CD8069504289900 (tray, SRFD9, retired) |
| Cores / Threads | 16 / 32 (Hyper-Threading) |
| Base / Max turbo frequency | 2.3 GHz / 3.7 GHz |
| Cache | 22 MB (1.375 MB per core) |
| TDP | 110 W (TCASE 81 °C) |
| Socket | FCLGA3647 (FC-LGA14B, 76.0 mm × 56.5 mm) |
| Memory | 6 channels DDR4-2667, up to 1 TB per socket, ECC RDIMM / LRDIMM / 3DS RDIMM, Optane persistent memory supported |
| Scalability | 4S (up to four sockets), 2 UPI links |
| PCIe | 3.0, 48 lanes |
| Speed Select | SST – Base Frequency: yes; SST – Performance Profile: no |
| AVX-512 | One FMA unit per core (Gold 5000-series rule; 5122 and 5222 are the exceptions with two) |
| Lithography | 14 nm |
| Launch | Q2 2019 |
| Status | Launched (servicing updates issued through 30 June 2025) |
| Embedded options | Available |
| Accelerators | AVX-512 (1 FMA), Intel DL Boost (VNNI), Intel RDT, Intel VMD, Run Sure |
Any FCLGA3647 board whose BIOS carries Cascade Lake microcode, in one-, two- and four-socket form. The deployments this bin exists for:
Two checks before ordering. Confirm the BIOS revision supports Cascade Lake, and confirm that the platform firmware actually exposes the SST–BF profile table if you intend to use it — the capability is in the processor, but the configuration interface is the vendor's. Boards that shipped before the 2019 launch usually need a firmware update before they will POST.
| Processor | Generation | Cores / Threads | Base / Max turbo | Cache | TDP | Memory support |
|---|---|---|---|---|---|---|
| Xeon Gold 5218N | 2nd Gen (Cascade Lake) | 16 / 32 | 2.3 / 3.7 GHz | 22 MB | 110 W | 6ch DDR4-2667 · 1 TB · 4S |
| Xeon Gold 5218 | 2nd Gen (Cascade Lake) | 16 / 32 | 2.3 / 3.9 GHz | 22 MB | 125 W | 6ch DDR4-2667 · 1 TB · 4S |
| Xeon Gold 5218R | 2nd Gen Refresh (Cascade Lake) | 20 / 40 | 2.1 / 4.0 GHz | 27.5 MB | 125 W | 6ch DDR4-2667 · 1 TB · 2S |
| Xeon Gold 5220 | 2nd Gen (Cascade Lake) | 18 / 36 | 2.2 / 3.9 GHz | 24.75 MB | 125 W | 6ch DDR4-2667 · 1 TB · 4S |
| Xeon Gold 5220T | 2nd Gen (Cascade Lake) | 18 / 36 | 1.9 / 3.9 GHz | 24.75 MB | 105 W | 6ch DDR4-2667 · 1 TB · 4S |
| Xeon Gold 6230 | 2nd Gen (Cascade Lake) | 20 / 40 | 2.1 / 3.9 GHz | 27.5 MB | 125 W | 6ch DDR4-2933 · 1 TB · 4S |
| Xeon Gold 6240 | 2nd Gen (Cascade Lake) | 18 / 36 | 2.6 / 3.9 GHz | 24.75 MB | 150 W | 6ch DDR4-2933 · 1 TB · 4S |
Against the 5218 the trade is exactly 200 MHz of turbo and 6 °C of case limit for 15 W, plus the embedded flag and SST–BF. Same cores, same cache, same base clock, same memory, same four-socket reach, same AVX-512. If the chassis or the facility is the constraint, the N is the better part; if peak burst speed is, the plain 5218 is.
Against the 5218R the refresh gives four more cores, 5.5 MB and a 4.0 GHz turbo, but it drops to two sockets and to 2.1 GHz base, and it carries neither the embedded flag nor SST–BF. Two-socket thread density points at the R; embedded and four-socket deployments point here.
Against the 5220S the step is two cores and 2.75 MB for 15 W more and a 400 MHz higher base clock, with no SST–BF. Against the 6230 it is four cores, 5.5 MB and 2933 memory for 15 W — and no embedded option. Against the 6240 two cores, 25 W, 300 MHz of base clock and the second AVX-512 FMA unit change hands, and against the 6252 or the 6254 it is a different thermal class entirely.
Every processor carries a 1-year warranty against defects in workmanship and function. Verified faulty units are replaced or refunded and we cover return shipping. Stock is held in both new and used condition; the condition you order is confirmed before dispatch, and used units are pull-tested stock rather than untested lots. Stock is held in our own warehouse and normally dispatches within three to seven working days, with tracking issued at dispatch.
Pricing is quoted per order because condition and quantity both move it. Send the quantity, the condition you need and the server or appliance the part is going into, and you will get a quote with lead time the same working day. If the design depends on SST–BF profiles, tell us the platform at the enquiry stage and we will confirm which sSpec build we are shipping before you commit.
Tell us who you are — we reply within one working day with a firm quote, stock status and lead time. Volume discounts available.
2.1 GHz / 8 Cores / 16 Threads / 11 MB Cache / 85 W / 78 °C TCASE
2.2 GHz / 8 Cores / 16 Threads / 11 MB Cache / 70 W / 91 °C TCASE
2.3 GHz / 10 Cores / 20 Threads / 13.75 MB Cache / 95 W / 91 °C TCASE
2.4 GHz / 10 Cores / 20 Threads / 13.75 MB Cache / 100 W / 2-Socket