基于OpenCL并行流动影响三维共晶生长多相场模拟
|
摘要
基于KKSO三维多元合金共晶多相场模型,研究了OpenCL+GPU软硬件体系结构的高性能计算方法,以CBr_4-C_2Cl_6为例,分别在AMD和NVIDIA 2种异构平台上通过多进程和多线程的并发执行实现了强迫对流下三维共晶组织的演化过程.结果表明:当计算规模相同时,分别在不同的平台上取得了一定的加速比.对并行算法在一定程度上进行优化,与CPU平台上的串行算法相比优化后的并行算法在异构平台上分别达到了20.2倍和23.6倍的加速比,大大提高了计算效率.同时,以其强大的浮点计算能力获取较为准确的模拟结果,达到计算效率和可移植性的双重需求,解决了传统求解相场模型存在的计算量大、效率低、限于定性研究等问题.
Based on KKSO three-dimensional multiphase-field model of multiple alloy eutectic, a high-performance computing method for OpenCL+GPU software and hardware architecture is studied. Taking CBr_4-C_2Cl_6 as an example, the evolution process of three-dimensional eutectic microstructure is implemented on condition of forced convection by means of multi-process and multi-threaded concurrent execution on two heterogeneous platforms of AMD and NVIDIA, respectively. The result shows that, on the same calculation scale, a certain acceleration ratio will be obtained respectively on the two platforms. The parallel algorithm optimized to a certain extent will be compared with the serial algorithm on CPU platform and the acceleration ratio of the former on heterogeneous platforms will be able to amount to 20.2 times and 23.6 times of the latter, respectively, improving greatly the calculation efficiency. At the same time, with powerful floating-point computation ability, to the more accurate simulation result will be acquired, achieving the double demand of computational efficiency and transplanting ability. The problems of traditional solution of phase-field model such as large computation, low efficiency and limited qualitative research will be resolved.
Based on KKSO three-dimensional multiphase-field model of multiple alloy eutectic, a high-performance computing method for OpenCL+GPU software and hardware architecture is studied. Taking CBr_4-C_2Cl_6 as an example, the evolution process of three-dimensional eutectic microstructure is implemented on condition of forced convection by means of multi-process and multi-threaded concurrent execution on two heterogeneous platforms of AMD and NVIDIA, respectively. The result shows that, on the same calculation scale, a certain acceleration ratio will be obtained respectively on the two platforms. The parallel algorithm optimized to a certain extent will be compared with the serial algorithm on CPU platform and the acceleration ratio of the former on heterogeneous platforms will be able to amount to 20.2 times and 23.6 times of the latter, respectively, improving greatly the calculation efficiency. At the same time, with powerful floating-point computation ability, to the more accurate simulation result will be acquired, achieving the double demand of computational efficiency and transplanting ability. The problems of traditional solution of phase-field model such as large computation, low efficiency and limited qualitative research will be resolved.
引文
[1] COCHRANE R F,GREER A L,ECKLER K,et al.Dendrite growth velocities in undercooled Ni-Si alloys [J]. Materials Science & Engineering A,1991,133(91):698-701.
[2] DGINIBRE M,AKAMATSU S,FAIVRE G.Experimental determination of the stability diagram of a lamellar eutectic growth front [J]. Physical Review E,1997,56(1):780-796.
[3] LU Y,BECKERMANN C,RAMIREZ J C.Three-dimensional phase-field simulations of the effect of convection on free dendritic growth [J].Journal of Crystal Growth,2005,280(1/2):320-334.
[4] 朱耀产,王锦程,杨根仓,等.共晶变速生长的多相场模拟 [J].金属学报,2007,43(2):194-198.
[5] 杨玉娟,严彪.CBr4-C2Cl6共晶合金生长的三维多相场模拟Ⅰ:模型建立及验证 [J].金属学报,2010,46(7):781-786.
[6] 杨玉娟,严彪.CBr4-C2Cl6共晶合金生长的三维多相场模拟 Ⅱ:层片间距对形态演化的影响 [J].金属学报,2010,46(7):787-793.
[7] 路阳,冯小静,冯力,等.强迫对流下共晶组织生长的相场法模拟 [J].兰州理工大学学报,2017,43(2):19-24.
[8] ZHANG A,GUO Z P,XIONG S M.Phase-field-lattice Boltzmann study for lamellar eutectic growth in a natural convection melt [J].China Foundry,2017,14(5):373-378.
[9] 朱昶胜,冯力,王智平,等.三维枝晶生长的相场法数值模拟研究 [J].物理学报,2009,58(11):8055-8061.
[10] KARMA A,RAPPEL W J.Quantitative phase-field modeling of dendritic growth in two and three dimensions [J].Physical Review E,1998,57(4):4323-4349.
[11] 朱昶胜,贾金芳,冯力,等.基于GPU的流动影响枝晶生长相场方法 [J].兰州理工大学学报,2015,41(3):19-25.
[12] KIM S G,KIM W T,SUZUKI T,et al.Phase-field modeling of eutectic solidification [J].Journal of Crystal Growth,2004,261(1):135-158.
[13] GASTER B,HOWES L,KAELI D R,et al.Heterogeneous computing with OpenCL:revised OpenCL [M].Morgan Kaufmann,2015.
[14] 刘文志,陈轶,吴长江.OpenCL异构并行计算 [M].北京:机械工业出版社,2016.
[15] 贾海鹏,张云泉,龙国平,等.基于OpenCL的拉普拉斯图像增强算法优化研究 [J].计算机科学,2012,39(5):271-277.
[16] 贾金芳.基于GPU并行的相场模型模拟研究 [D].兰州:兰州理工大学,2015.
[17] ODE M,GYOON K S,SUZUKI T.Mathematical modeling of iron and steel making processes.recent advances in the phase-field model for solidification [J].Isij International,2001,41(10):1076-1082.
[2] DGINIBRE M,AKAMATSU S,FAIVRE G.Experimental determination of the stability diagram of a lamellar eutectic growth front [J]. Physical Review E,1997,56(1):780-796.
[3] LU Y,BECKERMANN C,RAMIREZ J C.Three-dimensional phase-field simulations of the effect of convection on free dendritic growth [J].Journal of Crystal Growth,2005,280(1/2):320-334.
[4] 朱耀产,王锦程,杨根仓,等.共晶变速生长的多相场模拟 [J].金属学报,2007,43(2):194-198.
[5] 杨玉娟,严彪.CBr4-C2Cl6共晶合金生长的三维多相场模拟Ⅰ:模型建立及验证 [J].金属学报,2010,46(7):781-786.
[6] 杨玉娟,严彪.CBr4-C2Cl6共晶合金生长的三维多相场模拟 Ⅱ:层片间距对形态演化的影响 [J].金属学报,2010,46(7):787-793.
[7] 路阳,冯小静,冯力,等.强迫对流下共晶组织生长的相场法模拟 [J].兰州理工大学学报,2017,43(2):19-24.
[8] ZHANG A,GUO Z P,XIONG S M.Phase-field-lattice Boltzmann study for lamellar eutectic growth in a natural convection melt [J].China Foundry,2017,14(5):373-378.
[9] 朱昶胜,冯力,王智平,等.三维枝晶生长的相场法数值模拟研究 [J].物理学报,2009,58(11):8055-8061.
[10] KARMA A,RAPPEL W J.Quantitative phase-field modeling of dendritic growth in two and three dimensions [J].Physical Review E,1998,57(4):4323-4349.
[11] 朱昶胜,贾金芳,冯力,等.基于GPU的流动影响枝晶生长相场方法 [J].兰州理工大学学报,2015,41(3):19-25.
[12] KIM S G,KIM W T,SUZUKI T,et al.Phase-field modeling of eutectic solidification [J].Journal of Crystal Growth,2004,261(1):135-158.
[13] GASTER B,HOWES L,KAELI D R,et al.Heterogeneous computing with OpenCL:revised OpenCL [M].Morgan Kaufmann,2015.
[14] 刘文志,陈轶,吴长江.OpenCL异构并行计算 [M].北京:机械工业出版社,2016.
[15] 贾海鹏,张云泉,龙国平,等.基于OpenCL的拉普拉斯图像增强算法优化研究 [J].计算机科学,2012,39(5):271-277.
[16] 贾金芳.基于GPU并行的相场模型模拟研究 [D].兰州:兰州理工大学,2015.
[17] ODE M,GYOON K S,SUZUKI T.Mathematical modeling of iron and steel making processes.recent advances in the phase-field model for solidification [J].Isij International,2001,41(10):1076-1082.