Jump to content

ArrayFire

From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Qwertyus (talk | contribs) at 09:35, 23 September 2015 (Added {{refimprove}} tag to article (TW)). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.
ArrayFire
Company typePrivate
IndustryHigh-performance computing
Founded2007
Headquarters3423 Piedmont Rd NE
Suite 330
Atlanta, Georgia 30305
USA
Key people
John Melonakos (CEO)
Gallagher Pryor (CTO)
Scott Blakesless (VP Sales)
Pavan Yalamanchili (VP Product)
ProductsComputer software
Websitearrayfire.com

AccelerEyes, doing business as ArrayFire, is an American software company that develops programming tools for parallel computing and graphics on graphics processing unit (GPU) chipsets. Based in Atlanta, the company released Jacket,[1] a tool to compile MATLAB code for CUDA-enabled GPUs, in June 2008. In an October 2008 article,[2][dead link] the number of current Jacket users is said to be in the tens of thousands, and includes engineers and scientists in technology companies such as General Electric, Google, Toyota, and Morgan Stanley.

Company history

File:AccelerEyesLogo.png
Old company logo.

AccelerEyes was founded in June 2007 and produces software libraries for graphics processing unit (GPU) programming.[2] AccelerEyes has products for GPU computing in C, C++, Fortran, Python, and MATLAB.

Jacket is a software product that enables MATLAB code to run on the GPU, using CUDA or OpenCL.[3] Jacket builds high-level interfaces to remove the lower-level complexity of GPU programming.[4] This is accomplished by enabling MATLAB code to run on Nvidia, Advanced Micro Devices (AMD), and other GPUs coprocessors. Jacket contains a runtime to automatically transform M-code for CUDA-enabled or OpenCL-enabled GPUs.

ArrayFire is a software library that enables C, C++ and Fortran code to run on the GPU.[5] There are two versions available, one for CUDA GPUs and one for OpenCL devices. ArrayFire was developed with funding from DARPA, who uses it in its "Memex" dark web search software.[6]

See also

References

  1. ^ US application 2009141034 
  2. ^ a b "AccelerEyes GPU toolbox brings supercomputer power to PCs". Tech Journal South. 2008-10-20. Retrieved October 27, 2008.
  3. ^ "Jacket - GPU computational backend for MATLAB". AccelerEyes. 2012-02-02. Retrieved February 2, 2012.
  4. ^ "Jacket: The GPU Engine for MATLAB". Supercomputing Conference 2008 (SC08), the International Conference for High-Performance Computing, Networking, Storage and Analysis. 2008-10-01. Retrieved October 27, 2008. [dead link]
  5. ^ "ArrayFire - GPU library for C, C++, Fortran, and Python". AccelerEyes. 2012-02-02. Retrieved February 2, 2012.
  6. ^ "Watch out Google, DARPA just open sourced its tech". The Daily Star. 21 April 2015.