Skip to content

Latest commit

 

History

History
22 lines (22 loc) · 513 Bytes

README.md

File metadata and controls

22 lines (22 loc) · 513 Bytes

NNS-CUDA

Implement 14 different approaches to optimize Nearest Neighbor Search.

Detail

  • V0: CPU Linear Search
  • V1: GPU Distance Matrix
  • V2: GPU Thrust
  • V3: GPU Shared Memory Tree Reduction
  • V4: GPU AoS2SoA
  • V5: GPU Texture Memory
  • V6: GPU Constant Memory
  • V7: GPU Multiblock
  • V8: GPU Reduction of 4 GPUs
  • V9: GPU Full Loop Expansion
  • V10: CPU KD-Tree
  • V11: GPU KD-Tree
  • V12: CPU Octree
  • V13: GPU Octree

Run

$ nvcc -Xcompiler -fopenmp -arch=sm_70 main.cu -o main
$ ./main