應(yīng)用多GPU的可壓縮湍流并行計算
發(fā)布時間:2018-01-31 19:54
本文關(guān)鍵詞: CUDA 圖形處理器 湍流 并行計算 計算流體力學(xué) 出處:《國防科技大學(xué)學(xué)報》2015年03期 論文類型:期刊論文
【摘要】:利用CUDA Fortran語言發(fā)展了基于圖形處理器(GPU)的計算流體力學(xué)可壓縮湍流求解器。該求解器基于結(jié)構(gòu)網(wǎng)格有限體積法,空間離散采用AUSMPW+格式,湍流模型為k-ωSST兩方程模型,采用MPI實現(xiàn)并行計算。針對最新的GPU架構(gòu),討論了通量計算的優(yōu)化方法及GPU計算與PCIe數(shù)據(jù)傳輸、MPI通信重疊的多GPU并行算法。進(jìn)行了超聲速進(jìn)氣道及空天飛機(jī)等算例的數(shù)值模擬以驗證GPU在大網(wǎng)格量情況下的加速性能。計算結(jié)果表明:相對于Intel Xeon E5-2670 CPU單一核心的計算時間,單塊NVIDIA GTX Titan Black GPU可獲得107~125倍的加速比。利用四塊GPU實現(xiàn)了復(fù)雜外形1.34億網(wǎng)格的快速計算,并行效率為91.6%。
[Abstract]:A computational fluid dynamics compressible turbulence solver based on graphics processor (GPU) is developed by using CUDA Fortran language, which is based on the finite volume method of structured meshes. Spatial discretization is based on AUSMPW scheme, turbulence model is k- 蠅 SST two-equation model, and MPI is used to realize parallel computation. Aiming at the latest GPU framework. The optimization method of flux calculation, GPU calculation and PCIe data transmission are discussed. Multiple GPU parallel algorithm for overlapping MPI communication. Numerical simulations of supersonic inlet and aircrafts are carried out to verify the acceleration performance of GPU in the case of large mesh quantities. The results show that:. The computational time relative to the single core of Intel Xeon E5-2670 CPU. Single block NVIDIA GTX Titan Black. GPU can get a speedup ratio of 107 ~ 125.The fast calculation of 134 million mesh with complex shape is realized by using four GPU blocks. The parallel efficiency is 91.6.
【作者單位】: 國防科技大學(xué)航天科學(xué)與工程學(xué)院;
【基金】:國家自然科學(xué)基金資助項目(91016010,91216117)
【分類號】:TP338.6
【正文快照】: 隨著硬件性能的提高及編程技術(shù)的改進(jìn),圖形處理器(Graphical Processing Unit,GPU)加速器在高性能計算領(lǐng)域逐漸得到廣泛的應(yīng)用。在最新公布的超級計算機(jī)Top500名單中共有62套系統(tǒng)采用了加速器/協(xié)處理器,其中采用GPU加速器有46套,而在最新的Green500名單中前10位的超級計算機(jī)均
【共引文獻(xiàn)】
相關(guān)期刊論文 前10條
1 李映坤;韓s,
本文編號:1479868
本文鏈接:http://sikaile.net/kejilunwen/jisuanjikexuelunwen/1479868.html
最近更新
教材專著