Parallel performance of an IB-LBM suspension simulation framework

Mountrakis, Lampros; Lorenz, Eric; Malaspinas, Orestis; Alowayyed, Saad; Chopard, Bastien; Hoekstra, Alfons G.

doi:10.1016/j.jocs.2015.04.006

Cited by 64 publications

(59 citation statements)

References 28 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An important example of the interface-capturing approach is the immersed boundary method [75], which was widely used in the simulation of biomechanical problems [76][77][78][79][80][81]. In this immersed technique, the solid is discretized using a Lagrangian mesh ‡ that can move freely on top of a background Eulerian mesh that spans the whole computational domain.…”

Section: Fluid-structure Interaction Methodsmentioning

confidence: 99%

A hybrid variational-collocation immersed method for fluid-structure interaction using unstructured T-splines

Casquero

Liu

Bona-Casas

et al. 2015

Int. J. Numer. Meth. Engng

View full text Add to dashboard Cite

Summary We present a hybrid variational‐collocation, immersed, and fully‐implicit formulation for fluid‐structure interaction (FSI) using unstructured T‐splines. In our immersed methodology, we define an Eulerian mesh on the whole computational domain and a Lagrangian mesh on the solid domain, which moves arbitrarily on top of the Eulerian mesh. Mathematically, the problem reduces to solving three equations, namely, the linear momentum balance, mass conservation, and a condition of kinematic compatibility between the Lagrangian displacement and the Eulerian velocity. We use a weighted residual approach for the linear momentum and mass conservation equations, but we discretize directly the strong form of the kinematic relation, deriving a hybrid variational‐collocation method. We use T‐splines for both the spatial discretization and the information transfer between the Eulerian mesh and the Lagrangian mesh. T‐splines offer us two main advantages against non‐uniform rational B‐splines: they can be locally refined and they are unstructured. The generalized‐α method is used for the time discretization. We validate our formulation with a common FSI benchmark problem achieving excellent agreement with the theoretical solution. An example involving a partially immersed solid is also solved. The numerical examples show how the use of T‐junctions and extraordinary nodes results in an accurate, efficient, and flexible method. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Section: Fluid-structure Interaction Methodsmentioning

confidence: 99%

A hybrid variational-collocation immersed method for fluid-structure interaction using unstructured T-splines

Casquero

Liu

Bona-Casas

et al. 2015

Int. J. Numer. Meth. Engng

View full text Add to dashboard Cite

show abstract

“…We define FSI-related computation as the contributions of the immersed boundary and finite element models, along with the related MPI communication. Load balance challenges have been discussed in previous FSI frameworks, such as [5] and [17]. Figure 8 shows that average FSI-related computation for compound capsules remains at 80% efficiency at 64 processors.…”

Section: Resultsmentioning

confidence: 98%

“…Additionally, the small problem size limits parallel performance: at 64 cores, every capsule is shared by multiple tasks and the parallel overhead becomes large relative to the computation time. In contrast, [17] demonstrated superior strong scaling over a much larger problem size.…”

Section: Resultsmentioning

confidence: 99%

“…The overlapping halo is set to half of the support of the delta function (2 grid points). The capsule workload is distributed on two levels, by capsule and by vertex, similar to the approach in [17]. First, each capsule is owned by the task to which its center belongs and this task is responsible for computing forces from the finite element model on the capsule.…”

Section: Methodsmentioning

confidence: 99%

“…Further, it is not well understood how the addition of a second membrane representing the nucleus alters the computational cost or parallel performance of the fluid-structure interaction (FSI) simulation, as previous performance studies have focused on red blood cells (e.g., [5, 23, 17]). Intuitively, the addition of a second membrane may increase FSI- and capsule-related computation by as much as 100%.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Numerical simulation of a compound capsule in a constricted microchannel

Gounley

Draeger

Randles

2017

Procedia Computer Science

View full text Add to dashboard Cite

Simulations of the passage of eukaryotic cells through a constricted channel aid in studying the properties of cancer cells and their transport in the bloodstream. Compound capsules, which explicitly model the outer cell membrane and nuclear lamina, have the potential to improve computational model fidelity. However, general simulations of compound capsules transiting a constricted microchannel have not been conducted and the influence of the compound capsule model on computational performance is not well known. In this study, we extend a parallel hemodynamics application to simulate the fluid-structure interaction between compound capsules and fluid. With this framework, we compare the deformation of simple and compound capsules in constricted microchannels, and explore how deformation depends on the capillary number and on the volume fraction of the inner membrane. The computational framework’s parallel performance in this setting is evaluated and future development lessons are discussed.

show abstract

Parallelizing and optimizing large‐scale 3D multi‐phase flow simulations on the Tianhe‐2 supercomputer

Wang

et al. 2015

Concurrency and Computation

View full text Add to dashboard Cite

Summary The lattice Boltzmann method (LBM) is a widely used computational fluid dynamics method for flow problems with complex geometries and various boundary conditions. Large‐scale LBM simulations with increasing resolution and extending temporal range require massive https://en.wikipedia.org/wiki/High-performance_computing#High-performance computing (HPC) resources, thus motivating us to port it onto modern many‐core heterogeneous supercomputers like Tianhe‐2. Although many‐core accelerators such as graphics processing unit and Intel MIC have a dramatic advantage of floating‐point performance and power efficiency over CPUs, they also pose a tough challenge to parallelize and optimize computational fluid dynamics codes on large‐scale heterogeneous system. In this paper, we parallelize and optimize the open source 3D multi‐phase LBM code openlbmflow on the Intel Xeon Phi (MIC) accelerated Tianhe‐2 supercomputer using a hybrid and heterogeneous MPI+OpenMP+Offload+single instruction, mulitple data (SIMD) programming model. With cache blocking and SIMD‐friendly data structure transformation, we dramatically improve the SIMD and cache efficiency for the single‐thread performance on both CPU and Phi, achieving a speedup of 7.9X and 8.8X, respectively, compared with the baseline code. To collaborate CPUs and Phi processors efficiently, we propose a load‐balance scheme to distribute workloads among intra‐node two CPUs and three Phi processors and use an asynchronous model to overlap the collaborative computation and communication as far as possible. The collaborative approach with two CPUs and three Phi processors improves the performance by around 3.2X compared with the CPU‐only approach. Scalability tests show that openlbmflow can achieve a parallel efficiency of about 60% on 2048 nodes, with about 400K cores in total. To the best of our knowledge, this is the largest scale CPU‐MIC collaborative LBM simulation for 3D multi‐phase flow problems. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Parallel performance of an IB-LBM suspension simulation framework

Cited by 64 publications

References 28 publications

A hybrid variational-collocation immersed method for fluid-structure interaction using unstructured T-splines

A hybrid variational-collocation immersed method for fluid-structure interaction using unstructured T-splines

Numerical simulation of a compound capsule in a constricted microchannel

Parallelizing and optimizing large‐scale 3D multi‐phase flow simulations on the Tianhe‐2 supercomputer

Contact Info

Product

Resources

About