A hybrid parallel DEM approach with workload balancing based on HSFC

Cintra, Diogo Tenório; Willmersdorf, Ramiro Brito; Lyra, Paulo Roberto Maciel; Lira, William Wagner Matos

doi:10.1108/ec-01-2016-0019

Cited by 7 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We quantify the load-balancing cost and identify the best load-balancing schemes for different types of granular flows frequently simulated with the DEM. Instead of focusing on the load-balance algorithms, which have been studied and compared in the literature [10,11,13,18,[20][21][22][23]30], we study the weighting strategies and consistent load-balancing schemes in different scenarios that may occur in DEM simulations. This research is useful to the researchers interested in using the DEM simulations and of particle-based methods to optimize the computational costs of their parallel simulations.…”

Section: Problem Definitionmentioning

confidence: 99%

See 1 more Smart Citation

Load-Balancing Strategies in Discrete Element Method Simulations

Golshan

Blais

2021

Processes

View full text Add to dashboard Cite

In this research, we investigate the influence of a load-balancing strategy and parametrization on the speed-up of discrete element method simulations using Lethe-DEM. Lethe-DEM is an open-source DEM code which uses a cell-based load-balancing strategy. We compare the computational performance of different cell-weighing strategies based on the number of particles per cell (linear and quadratic). We observe two minimums for particle to cell weights (at 3, 40 for quadratic, and 15, 50 for linear) in both linear and quadratic strategies. The first and second minimums are attributed to the suitable distribution of cell-based and particle-based functions, respectively. We use four benchmark simulations (packing, rotating drum, silo, and V blender) to investigate the computational performances of different load-balancing schemes (namely, single-step, frequent and dynamic). These benchmarks are chosen to demonstrate different scenarios that may occur in a DEM simulation. In a large-scale rotating drum simulation, which shows the systems in which particles occupy a constant region after reaching steady-state, single-step load-balancing shows the best performance. In a silo and V blender, where particles move in one direction or have a reciprocating motion, frequent and dynamic schemes are preferred. We propose an automatic load-balancing scheme (dynamic) that finds the best load-balancing steps according to the imbalance of computational load between the processes. Furthermore, we show the high computational performance of Lethe-DEM in the simulation of the packing of 108 particles on 4800 processes. We show that simulations with optimum load-balancing need ≈40% less time compared to the simulations with no load-balancing.

show abstract

Section: Problem Definitionmentioning

confidence: 99%

“…LIGGGHTS uses a recursive multi-sectioning algorithm on a Cartesian simulation grid for load-balancing [20]. Cintra et al [21,22] used a recursive coordinate bisection in a simulation of hopper discharge and landslide using the DEMOOP software. The performances of different load-balancing algorithms were compared in other works [10,11,23].…”

Section: Introductionmentioning

confidence: 99%

Load-Balancing Strategies in Discrete Element Method Simulations

Golshan

Blais

2021

Processes

View full text Add to dashboard Cite

show abstract

“…Recursive coordinate bisection 32 and its variants are the most representative geometric partitioning methods. The other family of geometry‐based partitioning is the space‐filling curve 33–35 . The basic idea underlying this algorithm is to reduce multidimensional data to a one‐dimensional case for a uniform grid by determining the ordering of the space‐filling curve passing through nodes in the domain.…”

Section: Introductionmentioning

confidence: 99%

“…The ant colony optimization algorithm 48 was introduced as a domain decomposition technique for parallel finite element analysis, and the genetic algorithm was implemented in the same way 49 . The literature also focused on determining the optimum scheme with the best partition result and highest efficiency among different algorithms 3,35,50–52 …”

Section: Introductionmentioning

confidence: 99%

A sparse domain decomposition method for parallel computing of a four‐dimensional lattice spring model

Zhao

2021

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

In this work, an improved domain decomposition method is developed to address workload imbalance when implementing the parallel computing of a four‐dimensional lattice spring model (4D‐LSM) to solve problems in rock engineering on a large scale. A cubic domain decomposition scheme is adopted and optimized by a simulated annealing algorithm (SAA) to minimize the workload imbalance among subdomains. The improved domain decomposition method is implemented in the parallel computing of the 4D‐LSM. Numerical results indicate that the proposed domain decomposition method can further improve the workload balance among processors, which is helpful to supersede the limit of computational scale when solving large‐scale geotechnical problems and decrease the runtime of the parallel 4D‐LSM by at most 40% compared to the original cubic decomposition method. This shows the practicability of the proposed method in parallel computing. Two types of target functions of SAA are tested, and their influence on the performance of the parallel 4D‐LSM is investigated. Finally, a computational model with one billion particles for one actual engineering application of using 4D‐LSM is realized, and the result shows the advantages of parallel computing.

show abstract

“…Measurements of the load imbalance with different MPI/OpenMP configurations with up to 128 threads are presented [22]. Building on this work, Cintra et al [23,24] demonstrate a hybrid OpenMP/MPI parallelization implemented in DEMOOP.…”

Section: Introductionmentioning

confidence: 99%

A systematic comparison of runtime load balancing algorithms for massively parallel rigid particle dynamics

Eibl

Rüde

2019

Computer Physics Communications

View full text Add to dashboard Cite

As compute power increases with time, more involved and larger simulations become possible. However, it gets increasingly difficult to efficiently use the provided computational resources. Especially in particlebased simulations with a spatial domain partitioning large load imbalances can occur due to the simulation being dynamic. Then a static domain partitioning may not be suitable. This can deteriorate the overall runtime of the simulation significantly. Sophisticated load balancing strategies must be designed to alleviate this problem. In this paper we conduct a systematic evaluation of the performance of six different load balancing algorithms. Our tests cover a wide range of simulation sizes, and employ one of the largest supercomputers available. In particular we study the runtime and memory complexity of all components of the simulation carefully. When progressing to extreme scale simulations it is essential to identify bottlenecks and to predict the scaling behaviour. Scaling experiments are shown for up to over one million processes.The performance of each algorithm is analyzed with respect to the quality of the load balancing and its runtime costs. Additionally an applied test case is used to judge the applicability of the best algorithms in real world applications. For all tests, the waLBerla multiphysics framework is employed. processes [6,7]. One important aspect of this initial domain partitioning is to achieve an equal workload for all cores. However, since the simulated system is dynamic, and the particles may migrate between subdomains, the workload might be shifted during the simulation. This leads to load imbalances, that can slow down the whole simulation. To overcome this problem, the domain partitioning must be adapted dynamically throughout the simulation and/or the subdomains must be reassigned to different processes.Many simulation frameworks have therefore adopted load balancing and results are published for simulations of various sizes. Related WorkCompared to rigid body dynamics, molecular dynamics simulations differ in some aspects, however, the load balancing problem is closely related. Therefore we also consider methods proposed in the context of molecular dynamics here. A slightly dated but still relevant review of different methods suitable for load balancing can be found in [8]. Owen et al. [9] use load balancing based on the ParMetis [10] graph partitioning library to balance their combined FEM-DEM simulation. They use two applied test cases namely a 2D bucket filling and a 3D hopper filling example. Measurements with up to 6 cores are presented. Deng et al. [11]present a runtime load balancing approach for molecular dynamics simulations which deforms the domain partitioning at runtime. The initial rectangular grid is optimized by moving the corners of all subdomains individually in space to adjust to the simulation. Good quality of the partitioning is reported for an artificial checkerboard scenario with no acting forces. The load balancing improves the runtime performance but...

show abstract

A hybrid parallel DEM approach with workload balancing based on HSFC

Cited by 7 publications

References 17 publications

Load-Balancing Strategies in Discrete Element Method Simulations

Load-Balancing Strategies in Discrete Element Method Simulations

A sparse domain decomposition method for parallel computing of a four‐dimensional lattice spring model

A systematic comparison of runtime load balancing algorithms for massively parallel rigid particle dynamics

Contact Info

Product

Resources

About