MPI Parallelisation of 3D Multiphase Smoothed Particle Hydrodynamics

Numerical study of icebreaking process with two different bow shapes based on developed particle method in parallel scheme

“…Therefore, the best strategy for saving computing time is to reduce the computational complexity in numerical particle integration. This can be solved by multiple processors sharing the total numbers of particles, in other words, the domain decomposition algorithm (Cui et al, 2020). In order to achieve the parallelisation of level ice-ship interaction, the level ice (computing domain) discretised into numbers of particles are decomposed into np processors, for example 9 np  , as shown in Fig.…”

A Review of SPH Techniques for Hydrodynamic Simulations of Ocean Energy Devices

“…For instance, DualSPHysics adopted OpenMP as the strategy of CPUs acceleration [107]. Nonetheless, the speedup of OpenMP could be limited caused by the increasing costs of the overhead of data communication from the shared memory to threads (see, e.g., [262,264]). Another typical SMSs framework is Intel's Thread Building Blocks (TBB) that was developed by the C++ language for parallel programming on multi-core processors.…”

“…In addition to the workload, another drawback of MPI is the expensive cost to build a massive multi-processor cluster that is less accessible for ordinary SPH practitioners. Despite this, the superiorities of MPI are evident; the most important one is that MPI can realize large-scale SPH simulations by employing substantial processors up to thousands and even tens of thousands as reported in [264,266,267].…”

A GPU-Based δ-Plus-SPH Model for Non-Newtonian Multiphase Flows

“…Particle-based methods are often criticized for their high calculation costs [47,48]. Different parallel computing architectures such as MPI [28], OpenMP [49], OpenCL [50], and CUDA [51] are commonly used to alleviate this drawback. Among these parallel computing architectures, CUDA is the most popular framework that can take advantage of the GPU's computational horsepower.…”

“…This parallel computing architecture is relatively simple and widely adopted in SPH models [25,26]. The message-passing interface (MPI) can structure the CPUs of a multi-machine cluster into a multi-node framework and provides a feasible approach for massive-scale simulations [27,28]. Graphics processing units (GPUs) are specialized computer processors that are very effective at data-parallel computation-intensive tasks.…”