Header menu link for other important links
X
Effective mapping of an SPH algorithm on massively parallel GPU architecture
Published in Institute of Electrical and Electronics Engineers Inc.
2019
Pages: 41 - 45
Abstract
In the present study, the performance of a Lagrangian, mesh-free, particle-based method called Smoothed Particle Hydrodynamics (SPH) is investigated on a General Purpose Graphics Processing Unit (GPGPU) architecture. A one-to-one mapping of host (CPU) function to device (GPU) kernel is particularly used. A new methodology of sorting the evolution of spatio-temporal data of particles based on cells is tested on GPU for efficiency measures such as speedup, Dynamic Random Access Memory (DRAM) utilization, warp execution, occupancy of each kernel with different grids, block sizes, etc. Thread-divergence caused by spline and Wendland families of weighting functions has been studied. In SPH algorithm, an overall speedup of 15× was achieved on GPU. © 2018 IEEE.
Concepts (19)
  •  related image
    Computational fluid dynamics
  •  related image
    Computer graphics
  •  related image
    Computer graphics equipment
  •  related image
    Dynamic random access storage
  •  related image
    Hydrodynamics
  •  related image
    Mapping
  •  related image
    Memory architecture
  •  related image
    Parallel architectures
  •  related image
    Program processors
  •  related image
    Screening
  •  related image
    CUDA
  •  related image
    DYNAMIC RANDOM ACCESS MEMORY
  •  related image
    GENERAL PURPOSE GRAPHICS PROCESSING UNIT (GPGPU)
  •  related image
    MASSIVELY PARALLELS
  •  related image
    Particle methods
  •  related image
    PARTICLE-BASED METHODS
  •  related image
    Smoothed particle hydrodynamics
  •  related image
    SPATIO-TEMPORAL DATA
  •  related image
    Graphics processing unit