本文關(guān)鍵詞： OpenCL 泛Kriging算法 異構(gòu)計(jì)算　出處：《測(cè)繪科學(xué)》2017年05期 　論文類型：期刊論文

【摘要】：針對(duì)泛Kriging插值算法在大量數(shù)據(jù)處理時(shí)的高耗時(shí)問(wèn)題,該文從異構(gòu)平臺(tái)主機(jī)端與OpenCL設(shè)備端的交互方式入手,采用OpenCL異構(gòu)平臺(tái)開發(fā)語(yǔ)言進(jìn)行泛Kriging算法并行化實(shí)現(xiàn)研究,解決了其在進(jìn)行大數(shù)據(jù)量處理時(shí)數(shù)據(jù)存儲(chǔ)、數(shù)據(jù)交互、多設(shè)備調(diào)度等一系列問(wèn)題。在K20Xm平臺(tái)上使用不同的數(shù)據(jù)集和參數(shù)對(duì)并行泛Kriging算法進(jìn)行測(cè)試,實(shí)驗(yàn)結(jié)果表明:與Intel Xeon E5-2670CPU平臺(tái)相比,并行程序插值部分加速比達(dá)到40倍以上,整體并行程序加速比達(dá)到了18倍。
[Abstract]:In view of the high time consuming of the universal Kriging interpolation algorithm when processing a large amount of data, this paper starts with the interaction between the host side of the heterogeneous platform and the OpenCL device side. Using OpenCL heterogeneous platform development language to implement the parallel implementation of pan-#en1# algorithm, it solves the problem of data storage and data interaction in the process of large data processing. Using different data sets and parameters to test the parallel universal Kriging algorithm on K20Xm platform. The experimental results show that compared with the Intel Xeon E5-2670 CPU platform, the speedup of the interpolation part of the parallel program is more than 40 times, and the speedup ratio of the whole parallel program is 18 times higher than that of the Intel Xeon E5-2670 CPU platform.
【作者單位】： 電子科技大學(xué)資源與環(huán)境學(xué)院;電子科技大學(xué)大數(shù)據(jù)研究中心遙感所;Center
【基金】：教育部中央高校基本科研業(yè)務(wù)費(fèi)專項(xiàng)資金項(xiàng)目 中國(guó)科學(xué)院遙感與數(shù)字地球研究所項(xiàng)目 四川省應(yīng)急測(cè)繪保障與地質(zhì)災(zāi)害監(jiān)測(cè)工程技術(shù)研究中心開放研究基金項(xiàng)目(K2014B003) 國(guó)家博士后基金項(xiàng)目(2011M501400)
【分類號(hào)】：TP338.6
【正文快照】： 3.Center for ComputationTechnology,Louisiana State University,Baton Rouge,LA 70803,U.S.)0引言Kriging插值算法是一種線性無(wú)偏最優(yōu)插值算法,著重于權(quán)值系數(shù)的確定,以使得內(nèi)插函數(shù)處于最佳的插值狀態(tài)。與傳統(tǒng)的插值算法相比,該算法在計(jì)算估計(jì)值綜合考慮了待插值點(diǎn)空間位，

本文編號(hào)：1485143