Xeon Phi

Not to be confused with the ATI Xenos, Xenon, or the regular Intel Xeon.

Xeon Phi
Xeon Phi 5100 without heatsink
General information
Launched2010
Discontinued2020[1]
Marketed byIntel
Designed byIntel
Common manufacturer
  • Intel
Performance
Max. CPU clock rate1.053 GHz to 1.7 GHz
Cache
L1 cache32 KB per core
L2 cache512 KB per core
Architecture and classification
ApplicationSupercomputers
High-performance computing
Technology node45 nm transistors to 14 nm transistors (tri-gate)
MicroarchitectureLarrabee
Instruction setx86-16 (except coprocessor form factor), IA-32, x86-64[2]
Extensions
Physical specifications
Cores
  • 32-72
Memory (RAM)
  • Up to 384 GB and 16 GB
  • Up to DDR4 115.4 GB/s with ECC support
  • MCDRAM 400+ GB/s
Sockets
Products, models, variants
Core names
  • Knights Ferry
  • Knights Corner
  • Knights Landing
  • Knights Mill
  • Knights Hill
Model
    • Xeon Phi 3100
    • Xeon Phi 5100
    • Xeon Phi 7100
    • Xeon Phi 7200

Xeon Phi[3] is a discontinued series of x86 manycore processors designed and made by Intel. It was intended for use in supercomputers, servers, and high-end workstations. Its architecture allowed use of standard programming languages and application programming interfaces (APIs) such as OpenMP.[4][5]

Xeon Phi launched in 2010. Since it was originally based on an earlier GPU design (codenamed "Larrabee") by Intel[6] that was cancelled in 2009,[7] it shared application areas with GPUs. The main difference between Xeon Phi and a GPGPU like Nvidia Tesla was that Xeon Phi, with an x86-compatible core, could, with less modification, run software that was originally targeted to a standard x86 CPU.

Initially in the form of PCI Express-based add-on cards, a second-generation product, codenamed Knights Landing, was announced in June 2013.[8] These second-generation chips could be used as a standalone CPU, rather than just as an add-in card.

The Tianhe-2 supercomputer uses Xeon Phi processors.

In June 2013, the Tianhe-2 supercomputer at the National Supercomputer Center in Guangzhou (NSCC-GZ) was announced[9] as the world's fastest supercomputer (as of June 2023, it is No. 10[10]). It used Intel Xeon Phi coprocessors and Ivy Bridge-EP Xeon E5 v2 processors to achieve 33.86 petaFLOPS.[11]

The Xeon Phi product line directly competed with Nvidia's Tesla and AMD Radeon Instinct lines of deep learning and GPGPU cards. It was discontinued due to a lack of demand and Intel's problems with its 10nm node.[12]

  1. ^ Ian Cutress & Anton Shilov (7 May 2019). "The Larrabee Chapter Closes: Intel's Final Xeon Phi Processors Now in EOL". Retrieved 12 March 2020.
  2. ^ "Intel® Xeon Phi™ Coprocessor System Software Developers Guide" (PDF). Intel. 8 November 2012. p. 16. Archived (PDF) from the original on 21 April 2024. Retrieved 1 May 2024.
  3. ^ Cite error: The named reference IntelXeonPhiName was invoked but never defined (see the help page).
  4. ^ robert-reed (4 February 2013). "Best Known Methods for Using OpenMP on Intel Many Integrated Core (Intel MIC) Architecture". software.intel.com.
  5. ^ Jeffers, James; Reinders, James (1 March 2013). Intel Xeon Phi Coprocessor High Performance Programming. Morgan Kaufmann. ISBN 978-0124104143.
  6. ^ Hruska, Joel (8 May 2019). "Intel Quietly Kills Off Xeon Phi". ExtremeTech.
  7. ^ "Intel scraps graphics chip based on Larrabee". Reuters. 6 December 2009.
  8. ^ Sodani, Avinash; et al. (2016). "Knights Landing: Second-Generation Intel Xeon Phi Product". IEEE Micro. 36 (2): 34–46. doi:10.1109/MM.2016.25. S2CID 28837176.
  9. ^ "TOP500 - June 2013". TOP500. Retrieved 18 June 2013.
  10. ^ "June 2023 | TOP500 Supercomputer Sites". www.top500.org.
  11. ^ "Intel Powers the World's Fastest Supercomputer, Reveals New and Future High Performance Computing Technologies". Retrieved 21 June 2013.
  12. ^ W1zzard (24 July 2018). "Intel is Giving up on Xeon Phi - Eight More Models Declared End-Of-Life". TechPowerUp.{{cite web}}: CS1 maint: numeric names: authors list (link)