Limits...
Toward GPGPU accelerated human electromechanical cardiac simulations.

Vigueras G, Roy I, Cookson A, Lee J, Smith N, Nordsletten D - Int J Numer Method Biomed Eng (2013)

Bottom Line: Specifically, we port to the GPU a number of components of CHeart--a CPU-based finite element code developed for simulating multi-physics problems.Speedup of up to 72 × compared with SC and 2.6 × compared with MC was also observed for the PDE solve.Using the same human geometry, the GPU implementation of mechanics residual/Jacobian computation provided speedups of up to 44 × compared with SC and 2.0 × compared with MC.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, King's College London, UK.

Show MeSH

Related in: MedlinePlus

(a) Benchmark problem mesh. For this mesh, we have used the following resolutions: 0.2 mm ( ∼ 58 K DOFs), 0.1 mm ( ∼ 443 K DOFs), and 0.05 mm ( ∼ 3.5 M DOFs); (b) LV mesh. For this mesh, we have used the following resolutions: 0.5 mm ( ∼ 2.5 M DOFs) and the second mesh a resolution of 0.2 mm ( ∼ 19 M DOFs); (c) mechanics mesh—with 352 quadratic hexahedral elements and 555 nodes (3605 DOFs).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4016759&req=5

fig02: (a) Benchmark problem mesh. For this mesh, we have used the following resolutions: 0.2 mm ( ∼ 58 K DOFs), 0.1 mm ( ∼ 443 K DOFs), and 0.05 mm ( ∼ 3.5 M DOFs); (b) LV mesh. For this mesh, we have used the following resolutions: 0.5 mm ( ∼ 2.5 M DOFs) and the second mesh a resolution of 0.2 mm ( ∼ 19 M DOFs); (c) mechanics mesh—with 352 quadratic hexahedral elements and 555 nodes (3605 DOFs).

Mentions: In this paper, we focus on the solution of the monodomain problem on tetrahedral and hexahedral grids. Here, an approximation Ωh of Ω is constructed by merging finitely many, non-overlapping elements, τ, which assemble to form the mesh, Th(Ω) (see Figure 2), that is,


Toward GPGPU accelerated human electromechanical cardiac simulations.

Vigueras G, Roy I, Cookson A, Lee J, Smith N, Nordsletten D - Int J Numer Method Biomed Eng (2013)

(a) Benchmark problem mesh. For this mesh, we have used the following resolutions: 0.2 mm ( ∼ 58 K DOFs), 0.1 mm ( ∼ 443 K DOFs), and 0.05 mm ( ∼ 3.5 M DOFs); (b) LV mesh. For this mesh, we have used the following resolutions: 0.5 mm ( ∼ 2.5 M DOFs) and the second mesh a resolution of 0.2 mm ( ∼ 19 M DOFs); (c) mechanics mesh—with 352 quadratic hexahedral elements and 555 nodes (3605 DOFs).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4016759&req=5

fig02: (a) Benchmark problem mesh. For this mesh, we have used the following resolutions: 0.2 mm ( ∼ 58 K DOFs), 0.1 mm ( ∼ 443 K DOFs), and 0.05 mm ( ∼ 3.5 M DOFs); (b) LV mesh. For this mesh, we have used the following resolutions: 0.5 mm ( ∼ 2.5 M DOFs) and the second mesh a resolution of 0.2 mm ( ∼ 19 M DOFs); (c) mechanics mesh—with 352 quadratic hexahedral elements and 555 nodes (3605 DOFs).
Mentions: In this paper, we focus on the solution of the monodomain problem on tetrahedral and hexahedral grids. Here, an approximation Ωh of Ω is constructed by merging finitely many, non-overlapping elements, τ, which assemble to form the mesh, Th(Ω) (see Figure 2), that is,

Bottom Line: Specifically, we port to the GPU a number of components of CHeart--a CPU-based finite element code developed for simulating multi-physics problems.Speedup of up to 72 × compared with SC and 2.6 × compared with MC was also observed for the PDE solve.Using the same human geometry, the GPU implementation of mechanics residual/Jacobian computation provided speedups of up to 44 × compared with SC and 2.0 × compared with MC.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, King's College London, UK.

Show MeSH
Related in: MedlinePlus