Implementation and Optimization of Computer Simulations with GPU Acceleration for Sino-Atrial Nodes

doi:10.7652/xjtuxb201407011

您当前的位置：

首页 >

文章列表页 >

Implementation and Optimization of Computer Simulations with GPU Acceleration for Sino-Atrial Nodes

更新时间：2025-07-09

- Implementation and Optimization of Computer Simulations with GPU Acceleration for Sino-Atrial Nodes
- Vol. 48, Issue 7, Pages: 60-64(2014)
- 作者机构：
  
  西安交通大学电气工程学院,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb201407011
  CLC： TP391.9
- Online First：10 July 2014，
  
  Published：2014
- 稿件说明：
移动端阅览
Implementation and Optimization of Computer Simulations with GPU Acceleration for Sino-Atrial Nodes[J]. 2014, 48(7): 60-64.
DOI：

Implementation and Optimization of Computer Simulations with GPU Acceleration for Sino-Atrial Nodes[J]. 2014, 48(7): 60-64. DOI： 10.7652/xjtuxb201407011.

摘要

针对窦房结电生理计算机仿真运算量巨大、耗时长的问题

提出了基于高性能图形处理单元(GPU)实现并行计算及优化的方法。首先考虑窦房结细胞中央和边缘的差异

构建了一维非匀质窦房结组织模型; 利用算子分裂方法使模型的解算任务具备并行性。根据具体解算过程提出了三种并行化策略

并对其中耗时最短的策略从线程块设置、数据交换频率以及存储模式等方面进行了进一步优化。结果表明:对于500个细胞的仿真

CUDA程序较串行程序的执行时间下降了60%

进一步优化后

CUDA程序的执行时间可下降84%; 窦房结组织越大

GPU的加速效果越明显。结果验证了GPU加速解算方法可显著提高窦房结模型的解算速度

降低实际执行时间。

Abstract

A parallel computation method with its optimization based on graphic processing unit(GPU)is proposed to improve the problem that the electrophysiological simulations for sino-atrial nodes(SAN)needs large amount of calculation and time consumption. A one-dimensional inhomogeneous model for tissue of SAN is established by considering the difference of properties in the central and the edge of SAN cell. Then

the calculation of the model is parallelly performed by using the operator splitting method. Three parallelization strategies are put forward based on the calculation process. The strategy with the shortest running time is further optimized by considering the block size

the data transfer and data partitioning across various memories. Simulation results with 500 cells show that

compared with a serial program

the execution time of the non-optimized CUDA program decreases by about 60%

and the execution time further reduces to 84% when the CUDA program is optimized. It is also observed that the larger the tissue is

the more significant the acceleration effect of GPU is. It can be concluded that the proposed GPU accelerating method significantly speeds up the computation of SAN tissue model and decreases the execution time.

关键词

Keywords

references

赵丽丽, 张盛兵, 张萌, 等. 基于CUDA的高速FFT计算[J]. 计算机应用研究, 2011, 28(4): 1556-1559.

ZHAO Lili, ZHANG Shengbin, ZHANG Meng, et al. High performance FFT computation based on CUDA[J]. Application Research of Computers, 2011, 28(4): 1556-1559.

JAYSHREE P, JITENDRA K, MADHURA B A. GPGPU processing in CUDA architecture[J]. Advanced Computing: An International Journal, 2012, 3(1): 105-105.

MACEDONIA M. The GPU enters computing's mainstream[J]. Computer, 2003, 36(10): 106-108.

HOANGTRONG T M, WILLIAMS G S B, LEDERER J W, et al. GPU-enabled stochastic spatiotemporal model of rat ventricular myocyte calcium dynamics[J]. Biophysical Journal, 2011, 100(3): 557.

YU Rongdong, ZHANG Yin, ZHANG Sanyuan, et al. GPU accelerated simulation of cardiac activities[J]. Journal of Computers, 2010, 5(11): 1700-1704.

SATO D, XIE Yuanfang, WEISS J N, et al. Acceleration of cardiac tissue simulation with graphic processing units[J]. Medical Biological Engineering Computing, 2009, 47(9): 1011-1015.

POTSE M, DUBE B, RICHER J, et al. A comparison of monodomain and bidomain reaction-diffusion models for action potential propagation in the human heart[J]. IEEE Transactions on Biomedical Engineering, 2006, 53(12): 2425-2435.

金印彬, 杨琳, 张虹, 等. 二维心室肌中动作电位传导的数值算法研究[J]. 西安交通大学学报, 2004, 38(8): 851-854.

JIN Yinbin, YANG Lin, ZHANG Hong, et al. Numerical algorithm for conduction of action potential in two dimensional cardiac ventricle tissue[J]. Journal of Xi'an Jiaotong University, 2004, 38(8): 851-854.

GARNY A, KOHL P, HUNTER P J, et al. One-dimensional rabbit sinoatrial node models: benefits and limitations[J]. Journal of Cardiovascular Electrophysiology, 2003, 14(10): 121-132.

ZHANG Henggui, HOLDEN A V, KODAMA I, et al. Mathematical models of action potentials in the periphery and center of the rabbit sinoatrial node[J]. American Journal of Physiology: Heart and Circulatory Physiology, 2000, 279(1): 397-421.

ZHANG Hong, WANG Juan, WU Ruijuan, et al. Effects of atrium on the electrical activity of sino-atrial node[J]. Key Engineering Materials, 2011, 467: 742-747.

唐轶轩,吴俊敏,陈国良,等.并行片上网络仿真器ParaNSim的设计及性能分析.2012,46(2):24-30.[doi:10.7652/xjtuxb201202005]

张学一,付晓梅,周化锰.一种格约减辅助的协作多址接入方案.2012,46(4):100-105.[doi:10.7652/xjtuxb201204017]

刘海涛,王磊,赵万华.考虑模态特性的高速机床进给系统刚度匹配研究.2014,48(1):90-95.[doi:10.7652/xjtuxb2014 01015]

丁军平,蔡皖东.面向对等网络SEInR传播模型的理论分析.2012,46(2):31-37.[doi:10.7652/xjtuxb201202006]

刘学,焦淑红,司锡才.观测域求积分卡尔曼滤波的机载无源定位算法.2011,45(5):137.[doi:10.7652/xjtuxb201105 025]

张保,曹海军,董小社,等.面向图形处理器重叠通信与计算的数据划分方法.2011,45(4):1-4.[doi:10.7652/xjtuxb 201104001]

康雨,闫相国,郑崇勋,等.任意可分负载的多轮调度算法.2009,43(8):125-129.[doi:10.7652/xjtuxb200908025]

郭振宇,程博,叶敏,等.一种并行混沌差异演化算法.2007,41(3):299-302.[doi:10.7652/xjtuxb200703008]

王润新,刘进军.无须考虑电源接法和性质的三相电压型整流器平均模型.2009,43(12):81-85.[doi:10.7652/xjtuxb 200912017]

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Three-Phase Boost Rectifier Average Model Independent of Connection Mode and Waveform Properties of AC Sources

Inhibition of the Turbulence Induced by Phase-2 Early Afterdepolarization

Sensitivity Analysis of Foil Geometrical Parameters for the Third-Generation CO2 Gas Foil Bearings Based on Design of Experiments Method

Review of the Application of EBSILON Software in Combined Heat and Power System Optimization

I/O Performance Modeling and Prediction Method for Large-Scale Parallel Applications Using Machine Learning

Related Author

王润新

刘进军

赵丹

刘袁

Related Institution

No data

⁰