基于GPU加速的定向凝固相场模拟计算研究

doi:10.11896/j.issn.1002-137X.2015.07.005

计算机科学 ›› 2015, Vol. 42 ›› Issue (7): 19-21.doi: 10.11896/j.issn.1002-137X.2015.07.005

• 2014’全国理论计算机科学年会 • 上一篇下一篇

基于GPU加速的定向凝固相场模拟计算研究

胡延苏,高昂,王志军,慕德俊

长安大学电子与控制工程学院西安710064,西北工业大学电子信息学院西安710072,西北工业大学材料学院西安710072,西北工业大学自动化学院西安710072

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受中国博士后科学基金资助

GPU Application on the Phase-field Simulation

HU Yan-su, GAO Ang, WANG Zhi-jun and MU De-jun

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 相场法作为一种极具优势的微观组织数值模拟方法,已经在凝固微观组织演化机制的研究中得到了广泛应用。然而无论是从计算尺度还是微观组织演化时间上考虑,相场模拟计算量均非常大,对计算机有着非常高的要求。相对于传统的中央处理器(CPU)计算,图形处理器(GPU)计算是最近发展的一种高效计算手段。提出了一种基于GPU加速的定向凝固相场模拟计算策略,实现了大尺度条件下的定向凝固界面形态演化的加速计算。计算结果表明,对于单个计算机,GPU计算与CPU计算的加速比可以高达30余倍。GPU加速将为相场模拟的发展及应用带来新的契机。

关键词: 相场,定向凝固,图像处理单元,统一计算设备架构

Abstract: As a very preponderant microstructure numerical simulation method,the phase field method has been widely used in the research of solidification microstructure evolutionary mechanism.But whether the simulation scale or the evolution time is considered,there is huge computation which sets a high demand for the computer.Compared with the traditional CPU,GPU is much more efficient for the parallel computing.The paper proposed a GPU-based acceleration strategy to complete the directional solidification phase field simulation in large scale.The results show that the computation time can reduce 30 times compared with a single CPU,which brings a new opportunity for the phase field simulation in large scale.

Key words: Phase field,Directional solidification,Graphics processing unit(GPU),Compute unified device architecture(CUDA)

胡延苏,高昂,王志军,慕德俊. 基于GPU加速的定向凝固相场模拟计算研究[J]. 计算机科学, 2015, 42(7): 19-21. https://doi.org/10.11896/j.issn.1002-137X.2015.07.005

HU Yan-su, GAO Ang, WANG Zhi-jun and MU De-jun. GPU Application on the Phase-field Simulation[J]. Computer Science, 2015, 42(7): 19-21. https://doi.org/10.11896/j.issn.1002-137X.2015.07.005

参考文献

[1] Boettinger W J,Warren J A.Simulation of the Cell to Plane Front Transition during Directional Solidification at High Velo-city[J].J.Cryst.Growth,1999,200(3/4):583-591
[2] Wang J C,Zhang Y X,Yang Y J,et al.Phase field modeling for dendritic morphology transition and micro-segregation in multi-component alloys[J].Science in China Series E:Technological Sciences,2009,52(2):344-351
[3] Lan C W,Shih C J,Lee M H.Quantitative Phase Field Simulation of Deep Cells in Directional Solidification of an Alloy [J].Acta Mater,2005,53(8):2285-2294
[4] Gurevich S,Karma A,Plapp M,et al.Trivedi.Phase-Field Study of Three-Dimensional Steady-State Growth Shapes in Directional Solidification [J].Phys.Rev.E.,2010,81(1)
[5] Wang Y B,Wang Y X,Cheng Z H,et al.Phase-field Simulation of Interface Effect during Grain Nucleation of Solidification Processing [J].Rare Metal Materials and Engineering,2012,41(6):1045-1048
[6] Yamanaka A,Aoko T,Ogawa S,et al.GPU-accelerated phase-field simulation of dendritic solidification in a binary alloy [J].Journal of Crystal Growth,2010,318(1):40-45
[7] Takaki T,Yamanaka A,Nukada A,et al.Peta-scale phase-field simulation for dendritic solidification on the TSUBAME 2.0 supercomputer [C]∥Proceedings of 2011 International Conference for High Performance Computing,Networking,Storage and Analysis(SC ’11).2011
[8] Glaskowsky P N.NVIDIA’s Fermi:The First Complete GPU Computing Architecture[M]∥NVIDIA Corporation White Paper.2009
[9] Losert W,Shi B Q,Cummins H Z.Evolution of Dendritic Patterns during Alloy Solidification:Onset of the Initial Instability [J].Proc.Nat.Acad.Sci.USA.,1998,95(2):431-438

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于GPU加速的定向凝固相场模拟计算研究

GPU Application on the Phase-field Simulation

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0