Parallel Prefix Sum (Scan) with CUDA

763 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Accelerating Wright–Fisher Forward Simulations on the Graphics Processing Unit (2016, unknown): “Scan is similar in concept to reduce, wherein for each time step half of the binary operations are independent, but it is a more complex parallel algorithm that creates a running sum over the array rather than condensing the array to a single value [see Harris et al. (2007b)].”
KinectFusion: Real-time dense surface mapping and tracking (2011, methodology): “The symmetry of the system enables operations and memory to be saved and the final sum is obtained using a parallel tree-based reduction [13], to obtain the upper triangular component of the symmetric system.”
Designing efficient sorting algorithms for manycore GPUs (2009, methodology): “This approach has been implemented in CUDA by Harris et al. [24] and is publicly available as part of the CUDA Data-Parallel Primitives (CUDPP) library [25].”
SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression (2023, unknown): “We report both in Table 3, the “3-bit statistics“ entry corresponds to group size 16 with 3-bit statistics and “16-bit statistics“ stands for group size 16 with 16-bit statistics.”
State-of-the-art in heterogeneous computing (2010, methodology): “[75] proposed an implementation optimized for CUDA, which has been incorporated in the CUDA Data Parallel Primitives Library [74].”