As much as 56 Cores and 112 PCIe 5.0 Lanes
For all the singular focus that Intel has positioned on its client Core desktop CPU components in the previous couple of years, you might be forgiven for pondering that Intel has forgotten about their Xeon premium processor lineups for workstations. Between the de facto retirement of Intel’s desktop-grade Xeon W-1×00-series lineup, and the repeated delays of Intel’s current-generation large silicon components for servers, the Sapphire Rapids-based 4th Era Xeon Scalable sequence, there hasn’t been a lot noise from Intel within the workstation area in the previous couple of years. However now that Sapphire Rapids for servers has lastly launched, the logjam in Intel’s product roadmap has eventually cleared out, and Intel is lastly able to renew cascading their newest silicon into new workstation components.
This morning Intel is asserting their first top-to-bottom refresh of workstation components, the Xeon W-3400 and Xeon W-2400 sequence. Geared toward what Intel is broadly classifying because the Skilled Workstation and Mainstream Workstation markets respectively, these chip lineups are meant to be used in high-performance desktop workstation setups, significantly people who require extra CPU cores, extra PCIe lanes, extra reminiscence bandwidth, or a mixture of all three parts. Primarily based on the identical Sapphire Rapids silicon as Intel’s recently-launched server components, the brand new Xeon W SKUs will convey down many (however not all) of the options which have come to outline Intel’s modern server silicon, together with a brand new chipset (W790) and motherboards which might be extra appropriate to be used in high-performance workstations.
As with the brand new Xeon Scalable Elements, the massive three additions listed here are the shift to Intel’s Golden Cove CPU structure – with all of the IPC and clockspeed advantages that brings – together with the addition of assist for DDR5 reminiscence and PCIe 5 for I/O connectivity. All of which is a big improve over the combo of Cascade Lake and Ice Lake components that make up Intel’s earlier product stack. In the meantime in comparison with Intel’s present desktop processor lineup, these are all options that had been pioneered on Alder Lake (12th Gen Core) again in late 2021, the workstation-focused Xeon W components are going to be constructing issues out to a a lot bigger diploma.
Beginning on the prime, the Xeon W-3400 sequence (Sapphire Rapids-112L) will fluctuate from 12 to 56 cores, and all will embody 112 PCIe 5.0 lanes, assist for as much as 4 TB of DDR5-4800 reminiscence throughout eight reminiscence channels, ECC reminiscence (RDIMM-only), Intel vPro, and Intel Normal Manageability (ISM). 4 of the seven W-3400 SKUs (X-series) profit from unlocked multipliers and, as such, formally assist overclocking. In the meantime a step down from, the Xeon W-2400 sequence (Sapphire Rapids-64L), will supply between 6 and 24 CPU cores paired with a pared-down 64 lanes of PCIe 5.0 connectivity, assist for as much as 2 TB of DDR5-4800 reminiscence throughout 4 reminiscence channels, and all the remainder of the Xeon W trimmings similar to ECC reminiscence.
Intel Xeon Workstation Desktop Platforms | |||
AnandTech | 2021 | 2022 | 2023 |
Skilled Workstation | Xeon W 3300 (Ice Lake-64L) & Xeon W-3200 (Cascade Lake-64L) |
Xeon W-3400 (Sapphire Rapids-112L) |
|
Mainstream Workstation | Xeon W-2200 (Cascade Lake) | Xeon W-2400 (Sapphire Rapids-64L) |
|
Entry Workstation | Xeon W-1200 (Rocket Lake-S) |
twelfth Gen Core (Alder Lake-S) + W680 Chipset |
thirteenth Gen Core (Raptor Lake-S) + W680 Chipset |
The brand new Xeon W components shall be changing a mish-mash of various Xeon generations from Intel. Whereas Intel did launch some Ice Lake-based Xeon components in 2021 – the Xeon W-3300 household – these components had been a supplemental replace of kinds for Intel’s Xeon lineup, for particular prospects that wanted the additional CPU cores or PCIe bandwidth. For everybody else, the outgoing Xeon W product stack, the circa 2019 W-3200 and W-2200 households, have been primarily based on Intel’s Cascade Lake silicon – which itself was a modest replace to Intel’s Skylake components. So the significance of the launch of the Xeon W-3400/2400 sequence to Intel’s workstation lineup is difficult to overstate: this can be a main overhaul and improve of Intel’s Xeon workstation stack.
The brand new Xeon W components, in flip, shall be going up towards AMD’s Threadripper Professional 5000 WX components, that are primarily based on AMD’s Zen 3 structure. The latest Threadripper Professional components launched final spring, and AMD has primarily had the run of the market by way of CPU efficiency since then, because of a big benefit in core counts and IPC. Even with their new components, Intel technically nonetheless isn’t fully closing that core depend hole, however the increase in IPC, core counts, and clockspeeds ought to assist to degree the taking part in subject by way of general CPU efficiency – although by how a lot stays to be seen.
Intel Xeon W-3400 Collection: ‘Skilled’ Platform with As much as 56 Cores, 112 PCIe 5.0 lanes, and 8-Channel Reminiscence
Intel’s Xeon W-3400 and W-2400 sequence workstation processors are primarily based on Intel’s Golden Cove CPU structure, the identical structure as Intel’s Alder Lake (twelfth Gen) desktop processors. Representing the premier line-up from Intel’s 4th Gen Xeon Scalable Sapphire Rapids premium workstation choices, the W-3400 household has seven SKUs in complete. The Xeon W-3400 ranges from a modest 12-core/24-thread half (w5-3425) to a extremely anticipated 56-core/112-thread half, the flagship w9-3495X.
Intel Xeon W-3400 Collection (Sapphire Rapids-112L) | |||||||||
SKU | Cores/ Threads |
Base Freq (GHz) |
Turbo Freq (TB 2.0) |
Turbo Freq (TBM 3.0) |
PCI Lanes (Gen5) |
L3 Cache (MB) |
Unlocked (Perf Tuning) |
TDP (W) |
Worth (1KU) |
w9-3495X | 56/112 | 1.9 | 4.6 | 4.8 | 112 | 105 | Y | 350 | $5889 |
w9-3475X | 36/72 | 2.2 | 4.6 | 4.8 | 112 | 82.5 | Y | 300 | $3739 |
w7-3465X | 28/56 | 2.5 | 4.6 | 4.8 | 112 | 75 | Y | 300 | $2889 |
w7-3455 | 24/48 | 2.5 | 4.6 | 4.8 | 112 | 67.5 | N | 270 | $2489 |
w7-3445 | 20/40 | 2.6 | 4.6 | 4.8 | 112 | 52.5 | N | 270 | $1989 |
w5-3435X | 16/32 | 3.1 | 4.5 | 4.7 | 112 | 45 | Y | 270 | $1589 |
w5-3425 | 12/24 | 3.2 | 4.4 | 4.6 | 112 | 30 | N | 270 | $1189 |
For the Xeon W-3400 sequence specifically, these components are primarily based on Intel’s Sapphire Rapids Excessive Core Depend (XCC) silicon, which is at the moment utilized in Intel’s higher-end Xeon server components. The XCC silicon depends on 4 compute tiles, certain collectively utilizing Intel’s newest EMIB interconnect – a primary for a Xeon workstation processor.
The person tiles for a Sapphire Rapids XCC chip are all equivalent/symmetrical, so every tile gives 1 / 4 of the CPU cores, I/O, and reminiscence channels of your entire chip. As such, every tile can present as much as a most of 32 PCIe 5.0 lanes (112 complete on the w9-3495X), whereas every tile additionally contains as much as two reminiscence controllers offering eight-channel reminiscence throughout the W-3400 sequence.
Specializing in the top-end SKU of the Xeon W-3400 household, the Intel Xeon w9-3495X, it has comparable vibes to Intel’s earlier behemoth Xeon W-3175X, which was launched in 2019 and got here with official assist for overclocking. Just like the Skylake-based Xeon W-3175X, the most recent Xeon w9-3495X additionally has an unlocked multiplier for overclocking.
The Intel Xeon w9-3495X has 56 cores (for 112 threads), and in contrast to Intel’s desktop components, each final certainly one of these is a Efficiency (P) core. Additionally current is a complete of 105 MB of Intel’s Sensible L3 Cache, with official assist for eight-channels of DDR5-4800 ECC RDIMM reminiscence, with a most capability of as much as 4 TB.
Just like the server half it is primarily based on, w9-3495X has a relatively toasty TDP score, coming in at 350 Watts. And in apply, peak energy consumption is more likely to be a lot greater beneath full load with Intel’s Turbo Increase and Turbo Increase Max 3.0 applied sciences enabled, particularly on 56-unlocked cores. Though it has a base frequency on the 56 Golden Cove cores of 1.9 GHz, it has a turbo frequency of as much as 4.6 GHz, and because of Turbo Increase Max 3.0 (Intel’s favored core expertise), a handful of cores can increase additional to 4.8 GHz.
The opposite SKUs from the Xeon W-3400 household vary from 36-cores right down to 12-core choices, such because the w9-3475X (36C/72T) and the w5-3425 (12C/24T). Finally, all the Xeon W-3400 components supply the identical variety of DDR5 reminiscence channels and PCIe lanes, so what separates the completely different SKUs is CPU core counts, max reminiscence clockspeeds, L3 cache, and naturally, value.
In the meantime, as beforehand famous, 4 of the Xeon W-3400 SKUs – the w9-3495X, w9-3475X, w7-3465X, and the w5-3435X – are all “unlocked” processors. That is one thing Intel hasn’t provided on a Xeon W half in a couple of years and comes with some attention-grabbing ramifications. Moreover probably the most primary potential to change the clockspeed multipliers for the CPU, unlocked processors also can have their AVX and AMX offsets adjusted to maintain the processors from dropping fairly as a lot beneath heavy SIMD hundreds. Lastly, all of those components additionally supply some tuning choices for his or her mesh interconnects, although Intel hasn’t stated what exactly could be tweaked right here.
Costs on the Intel Xeon W-3400 household begin at $1189, with Intel offering pricing on a 1K per unit pricing (tray) and never individually bought retail SKUs. The Xeon w9-3495X has a 1KPU value of $5889, which makes the highest SKU and every subsequent W-3400 SKU dearer than the earlier technology of Xeon W-3300 chips, however they do include greater core counts, sooner turbo frequencies, extra L3 cache, and assist for DDR5-4800.
It’s value stating that each one of Intel’s W-3400 SKUs function assist for as much as 4TB of eight-channel DDR5-4800 ECC reminiscence, even the underside SKU, the w5-3425 (12C/24T). So there are alternatives within the Xeon product stack for programs that want an entire lot of DRAM, however not essentially a ton of CPU cores. Do notice, nevertheless, that really hitting 4TB requires utilizing 2 DIMMs per channel (DPC), which requires backing off to DDR5-4400 reminiscence speeds.
With 112 PCIe 5 lanes accessible from the CPU (and but extra from the chipset), the Xeon W-3400 chips can assist a relatively large variety of I/O units. This works out to seven discrete x16 graphics playing cards, or as much as 28 x4 high-speed storage units. This, together with core counts and reminiscence channels, is likely one of the major differentiators from the lower-tier Xeon W-2400 sequence – and needs to be a welcome growth for Intel platform customers who had been caught with a fraction of the I/O bandwidth on Intel’s earlier Xeon W components.
Curiously, 112 PCIe 5 lanes is definitely greater than Intel affords in its Sapphire Rapids server components. The Xeon Scalable lineup tops out at simply 80 lanes. This discrepancy comes from the truth that Intel solely enabled 5 of the 7 root ports for his or her server components, leaving an extra 2 ports (32 lanes) unused. Nonetheless because the workstation Sapphire Rapids components don’t must allocate any pins to supporting Intel’s multi-socket UPI hyperlinks, it might appear that Intel has as an alternative allotted these pins to carrying the extra PCIe lanes for the workstation components. It is value noting that Intel is utilizing the identical socket for each server and workstation chips right here – LGA 4677 – however with the pin modifications I would not count on them to be suitable.
In the meantime, in one other first for Intel, the corporate has stated that they will assist DDR5 XMP 3.0 reminiscence overclocking profiles for RDIMMs. The main points on this announcement are very scant, however at a excessive degree it’s going to give unlocked processor homeowners operating on W790 the choice of making an attempt to squeeze extra out of their reminiscence if they’ll. Usually talking, reminiscence overclocking and the rock-solid stability of RDIMMs are diametrically opposing targets, so it will likely be attention-grabbing to see how this performs out available in the market. The DRAM might find a way clock greater than simply DDR5-4800, however can the registered clock drivers (RCDs)?
As an apart, all of this speak explicitly round RDIMMs is intentional: in a giant change from earlier Xeon W platforms, the Sapphire Rapids Xeon workstation platforms is not going to assist UDIMMs. This can be a limitation of the DDR5 specification, which calls for various voltages for UDIMMs and RDIMMs respectively. Whereas UDIMMs take 5 volts, RDIMMs take 12 volts, rendering them incompatible. In the event you’ve ever had the possibility to see an DDR5 RDIMM in individual, you could have seen that they’re even keyed otherwise from UDIMMs, so they’re each bodily and electrically incompatible.
Finally, which means that customers will must pair these processors and W790 motherboards with dearer, albeit higher-quality ECC-enabled DDR5 RDIMMs. For dyed within the wool workstation customers that is unlikely to be a problem (or perhaps a distinction that will get seen), however anybody hoping to construct an HEDT-style system or low-end workstation on a budget goes to search out that the ultimate price ticket for a Xeon W system goes to be greater than what you might pull off with the W-3200/2200 sequence.
Accelerated Computing: AMX Makes the Minimize, Most Area-Particular Accelerators Do Not
For his or her Sapphire Rapids Xeon silicon and ensuing server components, Intel launched a slew of various acceleration blocks and different accelerator-related options. Between matrix extensions (AMX), varied area particular {hardware} acceleration blocks, and assist for Compute eXpress Hyperlink (CXL) for exterior accelerators, Intel ended up devoting a good bit of silicon to non-CPU duties. This has meant that for his or her Xeon Scalable server components specifically, Intel has opted (if not wanted) to lean on these accelerator options, with one DSA engine enabled in all the chips. Nonetheless, QAT, DLB, and IAA will not be supported. This in lieu of simply uncooked x86 CPU efficiency for differentiating the {hardware} from its predecessors and its competitors.
However for his or her workstation components, issues are just a little extra easy, for higher and for worse. In brief, not all of Intel’s accelerated computing options are being made accessible within the Xeon W-3400/2400 households. So let’s do a fast rundown of which of Sapphire Rapids extra esoteric options made the lower for Xeon W.
Maybe most critically of all, Intel’s Superior Matrix Extensions (AMX) did make the lower, and assist for them is totally current and enabled on the Xeon W-3400/2400 household. AMX is Intel’s matrix math execution block, and just like tensor cores and different varieties of matrix accelerators, these are ultra-high-density blocks for effectively executing matrix math. AMX isn’t a devoted accelerator, relatively it’s part of the CPU cores, with every core getting a block, which permits AMX code to be blended with x86 (and AVX) code, and can also be why Sapphire Rapids has detrimental clockspeed offsets for utilizing the ultra-dense code.
AMX is Intel’s play for the deep studying market, going above and past the throughput they’ll obtain as we speak with AVX-512 by utilizing even denser knowledge buildings. Whereas Intel has AMX-enabled GPUs (Intel Knowledge Middle Max GPU Collection) that transcend even this, for Sapphire Rapids Intel is trying to tackle the client phase that wants AI inference happening very near CPU cores, relatively than in a much less versatile, extra devoted accelerator. The brand new AMX models additionally assist Bfloat16, making certain that each tier of Intel’s accelerated computing blocks (AVX and AMX) assist this widespread mid-precision floating level format for deep studying.
Considered one of Sapphire Rapids’ new domain-specific {hardware} accelerator blocks, the Knowledge Streaming Accelerator (DSA), additionally made the lower. This block is for offloading/accelerating sure operations, similar to knowledge copies and easy computations similar to calculating CRC32s. The DSA block is accessible throughout all the Xeon W SKUs.
Nonetheless you will not discover point out of the remainder of Intel’s accelerator blocks, similar to Intel Dynamic Load Balancer (DLB), Intel In-Reminiscence Analytics Accelerator (IAA), and Intel QuickAssist Know-how (QAT). This even if these accelerators are all a part of the identical purposeful block on the Sapphire Rapids silicon. These different accelerator blocks are all primarily aimed toward servers, so it isn’t shocking to not see their inclusion, but it surely does imply anybody prototyping code for servers might want to take a look at on an precise Xeon Scalable in the event that they’re utilizing their options.
Lastly, CXL assist is absent from Intel’s Xeon W spec sheets, however Intel has confirmed to us that CXL is supported on each households. The built-on-top-of-PCIe customary for host-to-device connectivity has been within the wings for a couple of years now, and Sapphire Rapids is the primary Intel CPU platform to assist the expertise. Like a few of these different options, it’s primarily meant for servers, so there’s much less of an impetus to convey it to workstations. Nonetheless, Intel has enabled it for customers trying to leverage its performance.
Intel Xeon W-2400 Collection: As much as 24-Cores, 64 PCIe 5.0 lanes, For Mainstream Workstations
Dropping down a tier, now we have the Xeon W-2400 sequence (Sapphire Rapids-64L), which is designed as a ‘Mainstream’ workstation platform. Xeon W-2400 affords a bit greater than half as many PCIe lanes because the W-3400 SKUs, with 64 PCIe 5.0 lanes accessible, and the variety of reminiscence channels is lower in half as properly to 4 channels. As such, this implies costs are decrease on the W-2400 sequence than its beefier W-3400 counterparts, going as little as $359 for the entry-level Xeon w3-2423.
Intel Xeon W-2400 Collection (Sapphire Rapids-64L) | |||||||||
SKU | Cores/ Threads |
Base Freq (GHz) |
Turbo Freq (1T) |
Turbo Freq (TBM 3.0) |
PCI Lanes (Gen5) |
L3 Cache (MB) |
Unlocked (Perf Tuning) |
TDP (W) |
Worth (1KU) |
w7-2495X | 24/48 | 2.5 | 4.6 | 4.8 | 64 | 45 | Y | 225 | $2189 |
w7-2475X | 20/40 | 2.6 | 4.6 | 4.8 | 64 | 37.5 | Y | 225 | $1789 |
w5-2465X | 16/32 | 3.1 | 4.5 | 4.7 | 64 | 33.75 | Y | 200 | $1389 |
w5-2455X | 12/24 | 3.2 | 4.4 | 4.6 | 64 | 30 | Y | 200 | $1039 |
w5-2445 | 10/20 | 3.1 | 4.4 | 4.6 | 64 | 26.25 | N | 175 | $839 |
w3-2435 | 8/16 | 3.1 | 4.3 | 4.5 | 64 | 22.5 | N | 165 | $669 |
w3-2425 | 6/12 | 3.0 | 4.2 | 4.4 | 64 | 15 | N | 130 | $529 |
w3-2423 | 6/12 | 2.1 | 4.0 | 4.2 | 64 | 15 | N | 120 | $359 |
Total, the Xeon W-2400 sequence will vary from 6 cores as much as 24 cores. Intel is utilizing their Sapphire Rapids Medium Core Depend (MCC) silicon right here, which in contrast to the XCC silicon, is a conventional monolithic die. This implies no fancy EMIB packaging is required to construct the chip – as an alternative, Intel solely has to fab one relatively massive die.
On the top-end of the Xeon W-2400 lineup is the w7-2495X, which options 24-cores/48-threads, 45 MB of Intel Sensible L3 cache, and a TDP of 225 Watts. Intel additionally has three w5 sequence SKUs, and eventually the trio of w3 SKUs.
Like its expert-tier counterpart, the Xeon W-2400 sequence affords a constant reminiscence and I/O configuration throughout your entire lineup. This implies 64 lanes of PCIe 5 coming from the CPU, and 4 channels of DDR5 reminiscence, permitting for a most of two TB of reminiscence general. It it additionally value stating that solely the w5 and w7 SKUs supply full DDR-4800 reminiscence speeds; the w3 components are all capped at DDR4-4400. The silver lining? All SKUs drop to this pace in a 2 DPC configuration, so for those who had been trying to construct a 2 TB system for no matter motive, you will not get penalized.
Just like the Xeon W-3400 sequence, the W-2400 household additionally has a couple of unlocked X SKUs in its arsenal, together with the top-tier w7-2495X. Different SKUs with unlocked multipliers embody the w7-2475X with 20 cores and 37.5 MB of L3 cache, and two w5 SKUs (w5-2465X 16C/32T and w5-2455X 12C/24T). You will not discover any unlocked w3 components, nevertheless, as all three entry-level w3 SKUs are totally locked down.
Intel W790 Chipset: Helps each Xeon W-3400 and W-2400 Platforms
All of Intel’s Xeon W-3400 and W-2400 sequence SKUs profit from Intel vPro and Intel’s Normal Manageability (ISM) applied sciences. Each the Xeon W-2400 and W-3400 households are supported by the related W790 chipset, though CPU-specific options such because the variety of reminiscence channels and PCIe lanes accessible rely on the processor itself.
Among the fundamental options of the W790 chipset embody a Direct Media Interface (DMI) 4.0 x8 hyperlink between the processor and the chipset itself, in addition to as much as 16 PCIe 4.0 lanes and assist for as much as eight SATA 3.0 ports. W790 additionally helps as much as 5 USB 3.2 Gen2x2 (20Gbps) ports, contains an Intel Wi-Fi 6E PHY, and might assist 2.5 GbE controllers natively.
Though there’s no point out of latest motherboards, there are anticipated to be Intel W790 motherboards from distributors similar to ASUS, GIGABYTE, Supermicro, and ASRock. System integrators similar to Dell, Lenovo, and Supermicro are anticipated to take priority first in delivering options and programs earlier than DIY builders can get their palms on them.
The ASRock W790 WS motherboard
ASRock emailed us simply earlier than the launch to stipulate its W790 WS mannequin, with a 20+2-phase energy supply, twin 10 GbE controllers, and assist for as much as 2 TB of DDR5-4800 ECC RDIMMs throughout eight slots. Though this board helps each Xeon W-3400 and W-2400 processors, this board is just enabled for quad-channel reminiscence.
One thing value mentioning regarding the newest technology of motherboards is that W790 boards are more likely to value greater than the C621A-based boards that had been used to assist the Xeon W-3300 sequence (Ice Lake). It’s because W790 boards have 4 extra lanes of DDR5 reminiscence and 48 extra PCIe 5 lanes to account for. Whereas we count on to see completely different ranges of board designs with completely different slots and I/O configurations accessible sooner or later, Intel hasn’t specified if a few of these motherboards will assist each households, or if distributors will design particular boards across the particular person Xeon W-3400 and W-2400 sequence.
Intel’s Xeon W-3400 and W-2400 processors can be found to pre-order from trade companions, whereas programs deployments are anticipated someday in early March. Intel’s anticipated and beneficial pricing begins at $359 for the Xeon w3-2423 and goes as much as $5889 for the Xeon w9-3495X.