Analyzing and Modeling the Performance of the HemeLB Lattice-Boltzmann Simulation Environment
arXiv:1209.3972 · doi:10.1016/j.jocs.2013.03.002
Abstract
We investigate the performance of the HemeLB lattice-Boltzmann simulator for cerebrovascular blood flow, aimed at providing timely and clinically relevant assistance to neurosurgeons. HemeLB is optimised for sparse geometries, supports interactive use, and scales well to 32,768 cores for problems with ~81 million lattice sites. We obtain a maximum performance of 29.5 billion site updates per second, with only an 11% slowdown for highly sparse problems (5% fluid fraction). We present steering and visualisation performance measurements and provide a model which allows users to predict the performance, thereby determining how to run simulations with maximum accuracy within time constraints.
Accepted by the Journal of Computational Science. 33 pages, 16 figures, 7 tables
References in corpus (2)
Cited by in corpus (14)
- Massively Parallel Algorithms for the Lattice Boltzmann Method on Non-uniform Grids
- waLBerla: A block-structured high-performance framework for multiphysics simulations
- Computer simulations reveal complex distribution of haemodynamic forces in a mouse retina model of angiogenesis
- Choice of boundary condition for lattice-Boltzmann simulation of moderate Reynolds number flow in complex domains
- Impact of blood rheology on wall shear stress in a model of the middle cerebral artery
- Memory transfer optimization for a lattice Boltzmann solver on Kepler architecture nVidia GPUs
- Mesoscopic simulations at the physics-chemistry-biology interface
- Extreme-Scale Block-Structured Adaptive Mesh Refinement
- FabSim: facilitating computational research through automation on large-scale and distributed e-infrastructures
- Flexible composition and execution of high performance, high fidelity multiscale biomedical simulations
- Sparse geometries handling in lattice-Boltzmann method implementation for graphic processors
- An automated multiscale ensemble simulation approach for vascular blood flow
- Development and performance of a HemeLB GPU code for human-scale blood flow simulation
- Survey of Multiscale and Multiphysics Applications and Communities