|
gptkbp:instanceOf
|
Parallel Computing Concept
|
|
gptkbp:accessibleBy
|
Shared Memory
Constant Memory
Global Memory
Texture Memory
|
|
gptkbp:executedBy
|
gptkb:graphics_card
|
|
gptkbp:firstReleased
|
2007
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
CUDA Kernels
|
|
gptkbp:introduced
|
gptkb:NVIDIA
|
|
gptkbp:invokedFor
|
Device (Dynamic Parallelism)
Host CPU
|
|
gptkbp:launched
|
Host Code
|
|
gptkbp:launchedWithSyntax
|
<<<...>>> in CUDA C/C++
|
|
gptkbp:supportAtomicOperations
|
Yes
|
|
gptkbp:supportCoalescedMemoryAccess
|
Yes
|
|
gptkbp:supportDebugging
|
Yes
|
|
gptkbp:supportDeviceFunctions
|
Yes
|
|
gptkbp:supportDeviceLambdaFunctions
|
Yes (CUDA 8.0+)
|
|
gptkbp:supportDynamicParallelism
|
Yes
|
|
gptkbp:supportErrorHandling
|
Limited
|
|
gptkbp:supportGridExecution
|
Yes
|
|
gptkbp:supportGridStrideLoops
|
Yes
|
|
gptkbp:supportMultipleBlocks
|
Yes
|
|
gptkbp:supportMultipleReturnValues
|
No
|
|
gptkbp:supportMultipleThreads
|
Yes
|
|
gptkbp:supportParallelExecution
|
Yes
|
|
gptkbp:supportProfiling
|
Yes
|
|
gptkbp:supportRecursion
|
Limited
|
|
gptkbp:supportSharedMemory
|
Yes
|
|
gptkbp:supportStreamExecution
|
Yes
|
|
gptkbp:supportSynchronization
|
Yes
|
|
gptkbp:supportTemplateFunctions
|
Yes
|
|
gptkbp:supportThreadSynchronization
|
Yes
|
|
gptkbp:supportWarpLevelPrimitives
|
Yes
|
|
gptkbp:usedFor
|
gptkb:Machine_Learning
Image Processing
Deep Learning
High Performance Computing
Scientific Simulations
|
|
gptkbp:usedIn
|
NVIDIA CUDA Platform
|
|
gptkbp:writtenBy
|
gptkb:CUDA_C++
gptkb:CUDA_Fortran
gptkb:CUDA_C
|
|
gptkbp:bfsParent
|
gptkb:CUDA_Graph_API
gptkb:CUDA_Graphs
|
|
gptkbp:bfsLayer
|
7
|