Limits...
Parallel Optimization of 3D Cardiac Electrophysiological Model Using GPU.

Xia Y, Wang K, Zhang H - Comput Math Methods Med (2015)

Bottom Line: This imposes a big challenge to the traditional computation resources based on CPU environment, which already cannot meet the requirement of the whole computation demands or are not easily available due to expensive costs.GPU as a parallel computing environment therefore provides an alternative to solve the large-scale computational problems of whole heart modeling.Furthermore, several optimization strategies were proposed based on the features of the virtual heart model, which enabled a 200-fold speedup as compared to a CPU implementation.

View Article: PubMed Central - PubMed

Affiliation: School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China.

ABSTRACT
Large-scale 3D virtual heart model simulations are highly demanding in computational resources. This imposes a big challenge to the traditional computation resources based on CPU environment, which already cannot meet the requirement of the whole computation demands or are not easily available due to expensive costs. GPU as a parallel computing environment therefore provides an alternative to solve the large-scale computational problems of whole heart modeling. In this study, using a 3D sheep atrial model as a test bed, we developed a GPU-based simulation algorithm to simulate the conduction of electrical excitation waves in the 3D atria. In the GPU algorithm, a multicellular tissue model was split into two components: one is the single cell model (ordinary differential equation) and the other is the diffusion term of the monodomain model (partial differential equation). Such a decoupling enabled realization of the GPU parallel algorithm. Furthermore, several optimization strategies were proposed based on the features of the virtual heart model, which enabled a 200-fold speedup as compared to a CPU implementation. In conclusion, an optimized GPU algorithm has been developed that provides an economic and powerful platform for 3D whole heart simulations.

No MeSH data available.


Data structure comparison between CPU and GPU. (a) CPU with heterogeneous storage of model parameters; (b) GPU with homogeneous storage of model parameters reducing memory access time for each iteration of simulations.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4637086&req=5

fig1: Data structure comparison between CPU and GPU. (a) CPU with heterogeneous storage of model parameters; (b) GPU with homogeneous storage of model parameters reducing memory access time for each iteration of simulations.

Mentions: In the program of CPU version, we defined a class for each cell mesh, all parameters of cell, including action potential, current, and ionic concentration, were declared as member variables within the class, which was very convenient and clear for programming. But for GPU version, the data structure of cell class was very low in efficiency. The reason is that memory access in single thread programming is very different from that in multithread multiprogramming. In general, for the present mainstream GPU, one operation of memory reading will get continuous 128 bytes. Based on this fact, we constructed a continuous memory access for variables and parameters in multithread as shown in Figure 1.


Parallel Optimization of 3D Cardiac Electrophysiological Model Using GPU.

Xia Y, Wang K, Zhang H - Comput Math Methods Med (2015)

Data structure comparison between CPU and GPU. (a) CPU with heterogeneous storage of model parameters; (b) GPU with homogeneous storage of model parameters reducing memory access time for each iteration of simulations.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4637086&req=5

fig1: Data structure comparison between CPU and GPU. (a) CPU with heterogeneous storage of model parameters; (b) GPU with homogeneous storage of model parameters reducing memory access time for each iteration of simulations.
Mentions: In the program of CPU version, we defined a class for each cell mesh, all parameters of cell, including action potential, current, and ionic concentration, were declared as member variables within the class, which was very convenient and clear for programming. But for GPU version, the data structure of cell class was very low in efficiency. The reason is that memory access in single thread programming is very different from that in multithread multiprogramming. In general, for the present mainstream GPU, one operation of memory reading will get continuous 128 bytes. Based on this fact, we constructed a continuous memory access for variables and parameters in multithread as shown in Figure 1.

Bottom Line: This imposes a big challenge to the traditional computation resources based on CPU environment, which already cannot meet the requirement of the whole computation demands or are not easily available due to expensive costs.GPU as a parallel computing environment therefore provides an alternative to solve the large-scale computational problems of whole heart modeling.Furthermore, several optimization strategies were proposed based on the features of the virtual heart model, which enabled a 200-fold speedup as compared to a CPU implementation.

View Article: PubMed Central - PubMed

Affiliation: School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China.

ABSTRACT
Large-scale 3D virtual heart model simulations are highly demanding in computational resources. This imposes a big challenge to the traditional computation resources based on CPU environment, which already cannot meet the requirement of the whole computation demands or are not easily available due to expensive costs. GPU as a parallel computing environment therefore provides an alternative to solve the large-scale computational problems of whole heart modeling. In this study, using a 3D sheep atrial model as a test bed, we developed a GPU-based simulation algorithm to simulate the conduction of electrical excitation waves in the 3D atria. In the GPU algorithm, a multicellular tissue model was split into two components: one is the single cell model (ordinary differential equation) and the other is the diffusion term of the monodomain model (partial differential equation). Such a decoupling enabled realization of the GPU parallel algorithm. Furthermore, several optimization strategies were proposed based on the features of the virtual heart model, which enabled a 200-fold speedup as compared to a CPU implementation. In conclusion, an optimized GPU algorithm has been developed that provides an economic and powerful platform for 3D whole heart simulations.

No MeSH data available.