|
gptkbp:instanceOf
|
parallel computing model
|
|
gptkbp:API
|
gptkb:CUDA_Driver_API
gptkb:CUDA_Runtime_API
|
|
gptkbp:developedBy
|
gptkb:NVIDIA
|
|
gptkbp:documentation
|
NVIDIA CUDA Toolkit Documentation
|
|
gptkbp:enables
|
massively parallel computations
|
|
gptkbp:hasConcept
|
threads
grids
shared memory
kernels
global memory
thread blocks
device memory
host memory
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
CUDA programming model
|
|
gptkbp:introducedIn
|
2007
|
|
gptkbp:latestReleaseVersion
|
CUDA 12 (as of 2024)
|
|
gptkbp:platform
|
gptkb:NVIDIA_GPU
|
|
gptkbp:relatedTo
|
gptkb:cuBLAS
gptkb:OpenACC
gptkb:cuDNN
gptkb:DirectCompute
gptkb:OpenCL
PyTorch (via GPU support)
TensorFlow (via GPU support)
Thrust (library)
|
|
gptkbp:supports
|
SIMT (Single Instruction, Multiple Threads)
|
|
gptkbp:supportsLanguage
|
gptkb:Python_(via_libraries)
gptkb:Fortran
gptkb:C++
C
|
|
gptkbp:usedFor
|
general-purpose computing on GPUs
|
|
gptkbp:usedIn
|
deep learning
high-performance computing
scientific computing
image processing
|
|
gptkbp:website
|
https://developer.nvidia.com/cuda-zone
|
|
gptkbp:bfsParent
|
gptkb:SIMT_architecture
gptkb:Wen-mei_W._Hwu
|
|
gptkbp:bfsLayer
|
7
|