cuFFT

E758503

cuFFT is NVIDIA's GPU-accelerated Fast Fourier Transform library designed to perform high-performance FFT computations on CUDA-enabled devices.

Try in SPARQL Jump to: Statements Referenced by

Statements (50)

Predicate Object
instanceOf CUDA library
FFT library
GPU-accelerated library
apiLanguage C NERFINISHED
C++
developer NVIDIA NERFINISHED
distributionForm binary library
header files
documentation NVIDIA cuFFT Library User Guide NERFINISHED
feature asynchronous execution with CUDA streams
batched execution
in-place transforms
out-of-place transforms
plan-based execution model
stream support
includedWith NVIDIA CUDA Toolkit installation
license proprietary
optimizedFor NVIDIA GPU architectures NERFINISHED
partOf CUDA Toolkit NERFINISHED
primaryFunction Discrete Fourier Transform computation
Fast Fourier Transform computation
programmingModel CUDA NERFINISHED
requires CUDA driver NERFINISHED
CUDA runtime NERFINISHED
CUDA-capable GPU
supportsDataType double-precision complex
double-precision real
single-precision complex
single-precision real
supportsDimension 1D FFT
2D FFT
3D FFT
supportsHardware CUDA-enabled GPU
supportsPlatform Linux
Windows
macOS
supportsPrecision double precision
single precision
supportsTransformType batched FFT GENERATED
complex-to-complex FFT GENERATED
complex-to-real FFT GENERATED
multi-dimensional FFT GENERATED
real-to-complex FFT GENERATED
supportsVendor NVIDIA GPU NERFINISHED
useCase computational chemistry
computational physics
image processing
machine learning
signal processing
spectral analysis

Referenced by (3)

Full triples — surface form annotated when it differs from this entity's canonical label.

CUDA Fortran supports cuFFT
cuBLAS relatedLibrary cuFFT
Compute Unified Device Architecture includesLibrary cuFFT
subject surface form: CUDA