ArrayFire is a high performance software library for parallel computing with an easy-to-use API. Its array based function set makes parallel programming simple.
ArrayFire's multiple backends (CUDA, OpenCL and native CPU) make it platform independent and highly portable.
A few lines of code in ArrayFire can replace dozens of lines of parallel computing code, saving you valuable time and lowering development costs.
Build ArrayFire from source
To build ArrayFire from source, please follow the instructions on our wiki.
Download ArrayFire Installers
ArrayFire binary installers can be downloaded at the ArrayFire Downloads page.
Build
Tests
| Linux x86 |
|
|
| Linux Tegra |
|
|
| Windows |
|
|
| OSX |
|
|
Test coverage:
#include <arrayfire.h>
#include <cstdio>
using namespace af;
int main(int argc, char *argv[])
{
try {
// Select a device and display arrayfire info
int device = argc > 1 ? atoi(argv[1]) : 0;
af::setDevice(device);
af::info();
printf("Create a 5-by-3 matrix of random floats on the GPU\n");
array A = randu(5,3, f32);
af_print(A);
printf("Element-wise arithmetic\n");
array B = sin(A) + 1.5;
af_print(B);
printf("Negate the first three elements of second column\n");
B(seq(0, 2), 1) = B(seq(0, 2), 1) * -1;
af_print(B);
printf("Fourier transform the result\n");
array C = fft(B);
af_print(C);
printf("Grab last row\n");
array c = C.row(end);
af_print(c);
printf("Create 2-by-3 matrix from host data\n");
float d[] = { 1, 2, 3, 4, 5, 6 };
array D(2, 3, d, af::afHost);
af_print(D);
printf("Copy last column onto first\n");
D.col(0) = D.col(end);
af_print(D);
// Sort A
printf("Sort A and print sorted array and corresponding indices\n");
array vals, inds;
sort(vals, inds, A);
af_print(vals);
af_print(inds);
} catch (af::exception& e) {
fprintf(stderr, "%s\n", e.what());
throw;
}
}
You can find our complete documentation over here.
Quick links:
Contributions of any kind are welcome! Please refer to
this document
to learn more about how you can get involved with ArrayFire.
Citations and Acknowledgements
If you redistribute ArrayFire, please follow the terms established in
the license.
If you wish to cite ArrayFire in an academic publication, please use the
following reference:
Formatted:
Yalamanchili, P., Arshad, U., Mohammed, Z., Garigipati, P., Entschev, P.,
Kloppenborg, B., Malcolm, J. and Melonakos, J. (2015).
ArrayFire - A high performance software library for parallel computing with an
easy-to-use API. Atlanta: AccelerEyes. Retrieved from https://github.com/arrayfire/arrayfire
BibTeX:
@misc{Yalamanchili2015,
abstract = {ArrayFire is a high performance software library for parallel computing with an easy-to-use API. Its array based function set makes parallel programming simple. ArrayFire's multiple backends (CUDA, OpenCL and native CPU) make it platform independent and highly portable. A few lines of code in ArrayFire can replace dozens of lines of parallel computing code, saving you valuable time and lowering development costs.},
address = {Atlanta},
author = {Yalamanchili, Pavan and Arshad, Umar and Mohammed, Zakiuddin and Garigipati, Pradeep and Entschev, Peter and Kloppenborg, Brian and Malcolm, James and Melonakos, John},
publisher = {AccelerEyes},
title = {{ArrayFire - A high performance software library for parallel computing with an easy-to-use API}},
url = {https://github.com/arrayfire/arrayfire},
year = {2015}
}
ArrayFire development is funded by ArrayFire LLC and several third parties,
please see the list of acknowledgements for further details.