Device-Wide Primitives#
CUB device-level single-problem parallel algorithms:
cub::DeviceAdjacentDifferencecomputes the difference between adjacent elements residing within device-accessible memorycub::DeviceForprovides device-wide, parallel operations for iterating over data residing within device-accessible memorycub::DeviceHistogramconstructs histograms from data samples residing within device-accessible memorycub::DevicePartitionpartitions data residing within device-accessible memorycub::DeviceMergemerges two sorted sequences in device-accessible memory into a single onecub::DeviceMergeSortsorts items residing within device-accessible memorycub::DeviceRadixSortsorts items residing within device-accessible memory using radix sorting methodcub::DeviceReducecomputes reduction of items residing within device-accessible memorycub::DeviceRunLengthEncodedemarcating “runs” of same-valued items withing a sequence residing within device-accessible memorycub::DeviceScancomputes a prefix scan across a sequence of data items residing within device-accessible memorycub::DeviceSelectcompacts data residing within device-accessible memorycub::DeviceTopKfinds the largest (or smallest) K items from an unordered list residing within device-accessible memory
CUB device-level segmented-problem (batched) parallel algorithms:
cub::DeviceSegmentedSortcomputes batched sort across non-overlapping sequences of data residing within device-accessible memorycub::DeviceSegmentedRadixSortcomputes batched radix sort across non-overlapping sequences of data residing within device-accessible memorycub::DeviceSegmentedReducecomputes reductions across multiple sequences of data residing within device-accessible memorycub::DeviceCopyprovides device-wide, parallel operations for batched copying of data residing within device-accessible memorycub::DeviceMemcpyprovides device-wide, parallel operations for batched copying of data residing within device-accessible memory