Wikipedia

Single instruction, multiple threads

This is an old revision of this page, as edited by Wootery (talk | contribs) at 15:16, 4 September 2014 (Reduce implication that latency hiding is an integral part of SIMT). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Single instruction, multiple thread (SIMT) is an "execution model" and abstraction on top of the hardware paradigm, single instruction, multiple data, introduced by Nvidia:[1][2]

[The G80 Nvidia GPU architecture] introduced the single-instruction multiple-thread (SIMT) execution model where multiple independent threads execute concurrently using a single instruction.

SIMT is intended to limit instruction fetching overhead,[3] and is used in modern GPUs in combination with 'latency hiding' to enable high-performance execution despite considerable latency in memory-access operations.[4]


See also

References

  1. ^ "Nvidia Fermi Compute Arcitecture Whitepaper" (PDF). http://www.nvidia.com/. NVIDIA Corporation. 2009. Retrieved 2014-07-17. {{cite web}}: External link in |website= (help)
  2. ^ "NVIDIA Tesla: A Unified Graphics and Computing Architecture". http://www.ieee.org/. IEEE. 2008. p. 6 (Subscription required.). Retrieved 2014-08-07. {{cite web}}: External link in |website= (help)
  3. ^ Rul, Sean; Vandierendonck, Hans; D’Haene, Joris; De Bosschere, Koen (2010). An experimental study on performance portability of OpenCL kernels. Symp. Application Accelerators in High Performance Computing (SAAHPC).
  4. ^ "Advanced Topics in CUDA" (PDF). cc.gatech.edu. 2011. Retrieved 2014-08-28.


 

This computer science article is a stub. You can help Wikipedia by expanding it.

Retrieved from "https://en.wikipedia.org/w/index.php?title=Single_instruction,_multiple_threads&oldid=624167205"