Efficient and scalable algorithms for smoothed particle hydrodynamics on hybrid shared/distributed-memory architectures

DOI10.1137/140964266MaRDI QIDQ3454847zbMATH OpenOpenAlexFDO

Authors Pedro Gonnet

Publication date 27 November 2015

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1404.2303

zbMATH Keywords

multicoressimulationsmoothed particle hydrodynamicstask-based parallelism

Mathematics Subject Classification ID

Theory of matrix inversion and generalized inverses (15A09) Determinants, permanents, traces, other special matrix functions (15A15) Factorization of matrices (15A23) Probabilistic methods, particle methods, etc. for boundary value problems involving PDEs (65N75) Parallel numerical computation (65Y05) Particle methods and lattice-gas methods (76M28)

Abstract: This paper describes a new fast and implicitly parallel approach to neighbour-finding in multi-resolution Smoothed Particle Hydrodynamics (SPH) simulations. This new approach is based on hierarchical cell decompositions and sorted interactions, within a task-based formulation. It is shown to be faster than traditional tree-based codes, and to scale better than domain decomposition-based approaches on hybrid shared/distributed-memory parallel architectures, e.g. clusters of multi-cores, achieving a $40imes$ speedup over the Gadget-2 simulation code.

Recommendations

• scientific article; zbMATH DE number 2232687
• Efficient implementation of smoothed particle hydrodynamics (SPH) with plane sweep algorithm
• A new multi-resolution parallel framework for SPH
• A portable parallel particle program
• Calculating short range forces on a massively parallel computer: SPH on the Connection Machine

Cited in

(19)

• Vectorising the smooth particle hydrodynamics
• A portable parallel particle program
• Proposal for fast calculation of particle interactions in SPH simulations
• An efficient optimization of Hll method for the second generation of Intel Xeon Phi processor
• Auto-balancing algorithm for parallel SPH simulation of materials in extremes
• Fast and accurate Voronoi density gridding from Lagrangian hydrodynamics data
• Calculating short range forces on a massively parallel computer: SPH on the Connection Machine
• A CUDA-based implementation of an improved SPH method on GPU
• VecDualSPHysics: a vectorized implementation of smoothed particle hydrodynamics method for simulating fluid flows on multi-core processors
• Neighbour lists in smoothed particle hydrodynamics
• Nearest neighbor search algorithm and its parallel implementation in the SPH method
• Defining and optimizing algorithms for neighbouring particle identification in SPH fluid simulations
• scientific article; zbMATH DE number 1759203 (Why is no real title available?)
• scientific article; zbMATH DE number 1844561 (Why is no real title available?)
• Improving linked-lists using tree search algorithms for neighbor finding in variable-resolution smoothed particle hydrodynamics
• Euro-Par 2004 Parallel Processing
• Efficient implementation of smoothed particle hydrodynamics (SPH) with plane sweep algorithm
• scientific article; zbMATH DE number 2232687 (Why is no real title available?)
• ALARIC: an algorithm for constructing arbitrarily complex initial density distributions with low particle noise for SPH/SPMHD applications

Describes a project that uses

Uses Software

• deal.II
• MPI
• StarPU
• TREESPH
• CHARM++
• QUARK
• GADGET

This page was built for publication: Efficient and scalable algorithms for smoothed particle hydrodynamics on hybrid shared/distributed-memory architectures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3454847)