Intel has initiated its product discontinuance plan for its Xeon Phi 7200-series processors codenamed Knights Landing (KNL). The said CPUs are used primarily for HPC applications, including supercomputers. As it appears, demand for these chips is not exactly great, which is why Intel does not see any reasons to keep producing them. At the same time, the company will keep offering its codenamed Knights Mill (KNM) solutions for Deep Learning.

Intel announced product discontinuance plan for the Xeon Phi 7210, 7210F, 7230, 7230F, 7250, 7250F, 7290, and 7290F processors on Monday. The process of their phasing out is not going to take long. The company’s customers are told to place their final orders on these CPUs by August 31, 2018. Meanwhile, the final products will be delivered by July 19, 2019.

As usual, Intel explained that “market demand for the products has shifted to other Intel products,” indicating that its Xeon Phi 7200-series processors in LGA3647-1 form-factor might not exactly popular among target customers in the technical computing space. The said processors feature from 64 to 72 HPC-enhanced Atom Silvermont cores that process up to four threads of code simultaneously and are outfitted with 16 GB of high-bandwidth MCDRAM memory. As for throughput, socketed Xeon Phi KNL processors offer 3-3.4 TFLOPS of double precision floating point (FP64) performance.

Intel announced the EOL plan for HPC-oriented many-core Xeon Phi 7200-series “Knights Landing” CPUs about 11 months after it decided not to release its Xeon Phi 7200-series coprocessor cards. The latter only offered performance of up to 3 TFLOPS FP64 and featured considerably more limited I/O options than their socketed counterparts: they only had 16 PCIe 3.0 lanes (vs. 36 lanes on CPUs) and they did not support Intel’s OmniPath fabric (unlike the Xeon Phi 7200F-series CPUs) that is particularly useful for supercomputer deployments. Meanwhile, since socketed Knights Landing CPUs do not support multi-processor configurations, the coprocessors cards could be used for density, which is crucially important for HPC applications. As it appears, Intel first decided to EOL KNL cards and eliminate their density advantage, but now it decided to discontinue even the processors, possibly because of low interest from potential customers.

AVX-512 Support Propogation by Various Intel CPUs
  Xeon, Core X General Xeon Phi  
Skylake-SP AVX512BW
AVX512DQ
AVX512VL
AVX512F
AVX512CD
AVX512ER
AVX512PF
Knights Landing
Cannon Lake AVX512VBMI
AVX512IFMA
AVX512_4FMAPS
AVX512_4VNNIW
Knights Mill
Ice Lake AVX512_VNNI
AVX512_VBMI2
AVX512_BITALG
AVX512+VAES
AVX512+GFNI
AVX512+VPCLMULQDQ
AVX512_VPOPCNTDQ
Source: Intel Architecture Instruction Set Extensions and Future Features Programming Reference (pages 12 and 13)
Bold: Supported by Knights Landing

The discontinuance of Intel’s Xeon Phi 7200-series Knights Landing CPUs does not mean the end of the road for the product family in general. Intel will continue to offer its Xeon Phi 7235/7285/7295 “Knights Mill” CPUs optimized for Deep Learning applications, but featuring a lot of similarities with the Knights Landing. Nonetheless, with the cancellation of Knights Hill and obvious problems with Intel’s 10 nm process technology, it looks like the company will not have a competitive Xeon Phi product for the HPC market for quite some time.

Related Reading

Source: Intel

Comments Locked

23 Comments

View All Comments

  • Elstar - Tuesday, July 24, 2018 - link

    I wonder what this implies about DP versus SP workloads on HPC clusters. The fact that Knights Mill has half the DP performance (and double the SP performance) when compared to Knights Landing implies that Intel misjudged what HPC customers want/wanted.
  • PeachNCream - Tuesday, July 24, 2018 - link

    No support for Crysis meant nobody could benchmark it to determine the answer to the question, "Can it run Crysis?" so no corporation was going to touch Knights Landing with a ten foot pole.
  • HStewart - Tuesday, July 24, 2018 - link

    PHI were never consider as gaming card - very limited scientific use.
  • jordanclock - Tuesday, July 24, 2018 - link

    Woosh.
  • ImSpartacus - Wednesday, July 25, 2018 - link

    Yeah, I don't know what Intel was thinking with that.

    Do they think that any modern corporation would buy hardware that hasn't been tested on Crysis? The thought is ridiculous.
  • Kevin G - Tuesday, July 24, 2018 - link

    Deep learning applications generally don't need that high of precision. The improvements in single and half precision are what the deep learning niche wants and that is what they got in Knights Mill.
  • lefty2 - Wednesday, July 25, 2018 - link

    Knight Mill was dead on arrival. It only exists because Intel made a promise to a certain customer:

    https://www.servethehome.com/intel-xeon-phi-x205-s...

    "The Intel Knights Mill product is essentially being launched to fulfill a promise to a customer base that there would be a product, based on Xeon Phi, that would target deep learning training for a scale-out market."
  • Elstar - Wednesday, July 25, 2018 - link

    1) When Intel makes bespoke chips, they don't market them, therefore the x205 chips aren't for any particular customer.
    2) "a customer base" means just the opposite of "a certain customer". It means "most customers", i.e. the "base" of a pyramid where the top are niche uses/users and the base represents most uses most of the time.
  • Kevin G - Tuesday, July 24, 2018 - link

    The PCIe card version only offered marginal density improvements while removing one of the chips advantages: memory capacity via six channels of DDR4. The 16 GB of HMC was plenty for providing a high bandwidth cache but the working data sets are often far larger. Being able to host 384 GB of DDR4 on top of the 16 GB of HMC per node was plenty of room. Various deployments were leveraging four nodes in a 2U chassis which is in-line with how many double wide PCIe cards could fit into a similar chassis. The other difference is that the Xeon Phi nodes could also accept a small single slot PCIe card, providing more IO or another accelerator like a FPGA.

    The flip side is that that PCIe cards would have had to leverage external fabric but it could have been Omnipath with the Sky Lake Xeons or ordinary Omnipath PCIe card. Scaling outward would have been easier due to the fewer number of nodes but it made the individual nodes far more expensive. Memory capacity of a dual socket Sky Lake-SP system would have been roughly the same due to higher capacity DIMM support and being able to leverage two DIMMs per channel.

    Ultimately the loss of the PCIe based Xeon Phis wasn't a big deal for HPC. As noted in the article, the delays to 10 nm production kill off Knights Hill which was to be the next Xeon Phi chip. It would have made sense for Intel to release a Knights Corner refresh for HPC that leverages the same socket as Sky Lake-SP and its features. It would have been a simple injection that leverages platforms that were already in the pipeline to keep Intel in the game. Greater performance per node would have been handled by dual socket boards. While not major changes in the grand scheme of things, these would have permitted Intel to get a product out and maintain their foot hold in the market. Instead we have Knights Mill which changes up the vector unit for deep learning tuning.
  • babadivad - Tuesday, July 24, 2018 - link

    Larabee's baby.

Log in

Don't have an account? Sign up now