Cardiovascular flow simulation at extreme scale

Zhou, Min; Sahni, Onkar; Kim, Hyun Jin; Figueroa, C. Alberto; Taylor, Charles A.; Shephard, Mark S.; Jansen, Kenneth E.

doi:10.1007/s00466-009-0450-z

Cited by 39 publications

(22 citation statements)

References 37 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[27] This capability has been [28] extended to two-phase flows where we use the level set method to track the boundary between two immiscible fluids (either compressible or incompressible -to model bubble coalescence [29] and two-phase turbulence [9,10]). PHASTA can use anisotropically adapted unstructured grids and regular grids and its highly scalable performance on massively parallel computers has already been demonstrated (the code has shown good scaling out to 288 * 1024 IBM Blue Gene processors, at JUGENE, BG/P (Germany) [30] and more recently up to 768,000 cores on Mira at Argonne National Laboratory).…”

Section: Description Of the Simulation 21 Dns Solver Overviewmentioning

confidence: 94%

DNS of turbulent flow with hemispherical wall roughness

Mishra

Bolotnov

2015

Journal of Turbulence

View full text Add to dashboard Cite

The present study of the effect of roughness density on the mean flow turbulence parameters is motivated by the need for new generation of boundary conditions for multiphase computational multiphase fluid dynamics (CMFD) models applied to boiling flows. Effect of roughness element density on the turbulent flow in a channel is quantified through direct numerical simulations (DNSs). The Navier-Stokes equations are solved using finite element method and bubbles are approximated as rigid near-hemispherical obstacles at the wall. Six different cases were analysed including channel flow with smooth wall and channel flow with rough wall for five different bubble nucleation site densities. Friction factor and the law of the wall was calculated and compared with the previously published results. Existing correlations for nucleating bubble site density dependency on a wall heat flux were used to obtain a relation between the heat flux and the friction factor, leading to the law of the wall dependency on the heat flux. This separate effect study provides new guidelines on how the heat flux in subcooled boiling regime affects the turbulence behaviour near the wall and guides the computational fluid dynamics model development for boiling two-phase flows. IntroductionDirect numerical simulation (DNS) has become an important tool to help develop new models for computational multiphase fluid dynamics (CMFD).[1] The detailed understanding of the boiling process in turbulent flow is one of the most interesting challenge problems that are faced today. This work attempts to separate and quantify the effect of non-detached bubbles on the liquid flow turbulence during nucleate boiling. We do not consider energy equation, and we simplify the bubble distribution, size and shape to more effectively develop new hydrodynamic boundary conditions for CMFD.In the past decade, DNS and interface tracking has been extensively used to study bubble-turbulence interaction. Ilic et al. quantified the turbulent kinetic energy (TKE) balance in bubble-induced turbulence [2] and evaluated the energy spectra in bubble-driven liquid flows using DNS.[3] Note that in this scenario the bubble-induced turbulence was not superimposed on the existing single-phase turbulence. Turbulent bubbly channel flows were extensively studied by under low Reynolds number conditions. Two-phase numerical simulations of heat transport in bubbling turbulence has been recently studied by Lakkaraju et al.[7,8] To specifically quantify flow parameters and the effect of bubbles at different stages of boiling, a system of separate effect studies have been conducted by Bolotnov et al.[9,10]

show abstract

Section: Description Of the Simulation 21 Dns Solver Overviewmentioning

confidence: 94%

DNS of turbulent flow with hemispherical wall roughness

Mishra

Bolotnov

2015

Journal of Turbulence

View full text Add to dashboard Cite

show abstract

“…Since there is little computation performed during mesh adaptation relative to the substantial increase in communications required as the given mesh is distributed to more processors, the scaling decreases on high core counts (note that a strong scaling study is performed and therefore, the problem size is the same). However, the analysis have been shown to scale strongly with the similar amount of work load provided [21,41]. The fact that mesh modification routines are able to scale on bigger core count with more entities involved into communication supports the statement that it is likely to at a minimum provide the equivalent scaling with more work load on the same number of parts.…”

Section: Parallel Anisotropic Boundary Layer Adaptivity On a Heat Tramentioning

confidence: 63%

“…Such meshes can only be solved using massively parallel computers [19][20][21][22]. To effectively execute such simulations the mesh adaptation procedures must operate in parallel on the same computer as the flow solution using the same form of parallel decomposition which is commonly represented as a partitioned mesh [20,21,23].…”

Section: Introductionmentioning

confidence: 99%

Parallel Adaptive Boundary Layer Meshing for CFD Analysis

Ovcharenko

Chitale

Sahni

et al. 2013

Proceedings of the 21st International Meshing Roundtable

Self Cite

View full text Add to dashboard Cite

Summary. This paper describes a parallel procedure for anisotropic mesh adaptation with boundary layers for use in scalable CFD simulations. The parallel mesh adaptation algorithm operates with local mesh modification operations developed for both unstructured and boundary layer parts of the mesh. The adaptive approach maintains layered elements near the viscous walls and accounts for the mesh modification operations that are carried out in parallel on a distributed mesh. In the process mesh relationships and approximations with respect to curved complex 3D geometries of interest are properly maintained. The parallel mesh adaptation procedures are applied to two problems: a heat transfer manifold and a scramjet engine.

show abstract

“…PHASTA [14] is a parallel, hierarchic, high-order accurate, adaptive, stabilized finite element solver for transient analysis of incompressible and compressible flows using advanced anisotropic adaptive algorithms and numerical models of flow physics. Its highly scalable performance on massively parallel computers has already been demonstrated on a variety of supercomputers such as Cray and IBM BlueGene [15] [16]. …”

Section: Description Of Our Approachmentioning

confidence: 97%

In-situ visualization and computational steering for large-scale simulation of turbulent flows in complex geometries

Hong

Rasquin

Fang

2014

2014 IEEE International Conference on Big Data (Big Data)

View full text Add to dashboard Cite

Large-scale simulations conducted on supercomputers such as leadership-class computing facilities allow researchers to simulate and study complex problems with high fidelity, and thus have become indispensable in diverse areas of science and engineering. These high-fidelity simulations generate vast amount of data which is becoming more and more difficult to transform into knowledge using traditional visual analysis approaches. For instance, there are tremendous challenges in analyzing big data produced by high-fidelity simulations in order to gain meaningful insight into complex phenomena such as turbulent two-phase flows. The traditional workflow, which consists in conducting simulations on supercomputers and recording enormous raw simulation data to disk for further post-processing and visualization, is no longer a viable approach due to prohibitive cost of disk access and considerable amount of time spent on data transfer. Visual Analytics approaches for big data have to be researched and employed to address the problem of knowledge discovery from such large-scale simulations. One approach to tackle this issue is to couple a numerical simulation with in-situ visualization so that the post-processing and visualization occurs while the simulation is running. This in-situ approach minimizes data storage by extracting and visualizing important features of the data directly within the simulation without saving the raw data to disk. In addition, in-situ visualization allows users to steer the simulation by adjusting input parameters while the simulation is ongoing. In this paper, we present our approach for in-situ visualization of simulation data generated by massively parallel finite-element computational fluid dynamics solver (PHASTA) instrumented and linked with ParaView Catalyst. We demonstrate our in-situ visualization and simulation steering capability with a fully resolved turbulent flow through 2×2 reactor subchannel complex geometry. In addition, we present results from our insitu visualization for turbulent flow simulations conducted on the supercomputers Cray XK7 "Titan" at Oak Ridge National Laboratory and IBM BlueGene/Q "Mira" at Argonne National Laboratory up to 32,768 cores and examine the overhead of insitu visualization and its effect on code performance.

show abstract

Cardiovascular flow simulation at extreme scale

Cited by 39 publications

References 37 publications

DNS of turbulent flow with hemispherical wall roughness

DNS of turbulent flow with hemispherical wall roughness

Parallel Adaptive Boundary Layer Meshing for CFD Analysis

In-situ visualization and computational steering for large-scale simulation of turbulent flows in complex geometries

Contact Info

Product

Resources

About