面向眾核加速器的異構(gòu)OpenCL代碼生成與優(yōu)化方法研究
本文選題:OpenCL 切入點:異構(gòu) 出處:《西華師范大學》2017年碩士論文 論文類型:學位論文
【摘要】:近年來,因為功耗的不斷增加、互連線延遲的限制以及設(shè)計復(fù)雜度的與日俱增,處理器性能的提高受到限制。傳統(tǒng)的單核體系結(jié)構(gòu)已經(jīng)難以滿足市場對性能的需求。隨著集成電路工藝的巨大發(fā)展,使得在單芯片上集成多個處理器核心來完成更復(fù)雜、大型的計算任務(wù)成為可能,處理器實現(xiàn)了從單核到多核以及眾核的轉(zhuǎn)變。但是,通過不斷增加同類型處理器核來提升性能同樣是存在瓶頸的。CPU的內(nèi)核數(shù)量在到達極限值后將無法再通過增加處理器核數(shù)來提升性能。為了進一步增強計算能力,硬件設(shè)計呈現(xiàn)出異構(gòu)化的趨勢。然而,由于底層的異構(gòu)以及多級存儲層次,異構(gòu)系統(tǒng)中的編程難問題成為制約異構(gòu)系統(tǒng)發(fā)展的瓶頸之一。因此,OpenCL作為首個異構(gòu)并行編程框架受到了人們越來越多的青睞。作為異構(gòu)計算的開放標準,OpenCL已經(jīng)得到了眾多廠商的大力支持,為異構(gòu)系統(tǒng)提供了一個免費的、開放的通用標準。為了實現(xiàn)核心處理器的自主化,我國自主研制的“神威·太湖之光”超級計算機,采用了國產(chǎn)片上異構(gòu)眾核處理器SW26010。為了降低程序員的編程難度、同時提高軟件的移植效率,本文設(shè)計并實現(xiàn)了支持國產(chǎn)SW26010眾核處理器的OpenCL編譯系統(tǒng),并且對OpenCL優(yōu)化方法進行了研究。本文的創(chuàng)新點主要包含以下幾部分:(1)基于OpenCL的編程框架,結(jié)合國產(chǎn)眾核處理器的微結(jié)構(gòu)特征,本文提出了OpenCL平臺模型、內(nèi)存模型和執(zhí)行模型向SW26010眾核處理器的映射關(guān)系。(2)針對硬件結(jié)構(gòu)的特征,本文提出了面向眾核加速器的線程合并、數(shù)據(jù)布局等OpenCL優(yōu)化方法。
[Abstract]:In recent years, because of the increasing power consumption, the limitation of interconnect delay and the increasing design complexity, The improvement of processor performance is limited. Traditional single-core architecture has been unable to meet the market demand for performance. With the rapid development of integrated circuit technology, it is more complicated to integrate multiple processor cores on a single chip. Large computing tasks are possible, with processors making the transition from single-core to multi-core and multi-core. Improving performance by increasing the number of cores of the same type of processors, which are also bottleneck. CPUs will not be able to improve performance by increasing the number of processor cores after reaching the limit. Hardware design shows a trend of isomerization. However, due to the underlying heterogeneity and multilevel storage levels, The difficulty of programming in heterogeneous systems has become one of the bottlenecks restricting the development of heterogeneous systems, so OpenCL as the first heterogeneous parallel programming framework has been more and more popular. As an open standard of heterogeneous computing, OpenCL has been acquired. With the strong support of many manufacturers, Provides a free, open and universal standard for heterogeneous systems. In order to realize the autonomy of the core processor, our country has developed our own "Shenwei Taihu Light" supercomputer. In order to reduce the programming difficulty of programmer and improve the efficiency of software transplantation, this paper designs and implements the OpenCL compiler system which supports the domestic SW26010 multicore processor. The innovation of this paper mainly includes the following parts: 1) the programming framework based on OpenCL, combined with the microstructural characteristics of the domestic multi-core processor, this paper proposes the OpenCL platform model. The mapping relationship between memory model and execution model to SW26010 multi-core processor. (2) aiming at the characteristics of hardware structure, this paper proposes a OpenCL optimization method such as thread merging, data layout and so on, which is oriented to the multi-kernel accelerator.
【學位授予單位】:西華師范大學
【學位級別】:碩士
【學位授予年份】:2017
【分類號】:TP332
【參考文獻】
相關(guān)期刊論文 前8條
1 Haohuan FU;Junfeng LIAO;Jinzhe YANG;Lanning WANG;Zhenya SONG;Xiaomeng HUANG;Chao YANG;Wei XUE;Fangfang LIU;Fangli QIAO;Wei ZHAO;Xunqiang YIN;Chaofeng HOU;Chenglong ZHANG;Wei GE;Jian ZHANG;Yangang WANG;Chunbo ZHOU;Guangwen YANG;;The Sunway Taihu Light supercomputer:system and applications[J];Science China(Information Sciences);2016年07期
2 鄭方;許勇;李宏亮;謝向輝;陳左寧;;一種面向高性能計算的自主眾核處理器結(jié)構(gòu)[J];中國科學:信息科學;2015年04期
3 蔡軍;許麗人;申曉瑩;;大氣環(huán)境仿真的工程化應(yīng)用研究[J];系統(tǒng)仿真學報;2015年01期
4 劉穎;呂方;王蕾;陳莉;崔慧敏;馮曉兵;;異構(gòu)并行編程模型研究與進展[J];軟件學報;2014年07期
5 楊海平;沈占鋒;駱劍承;吳煒;;海量遙感數(shù)據(jù)的高性能地學計算應(yīng)用與發(fā)展分析[J];地球信息科學學報;2013年01期
6 孟小峰;慈祥;;大數(shù)據(jù)管理:概念、技術(shù)與挑戰(zhàn)[J];計算機研究與發(fā)展;2013年01期
7 魏敏;王彬;孫婧;谷軍霞;洪文董;;“天河一號”系列超級計算機系統(tǒng)氣象領(lǐng)域適用性分析[J];氣象科技進展;2012年01期
8 李喬;鄭嘯;;云計算研究現(xiàn)狀綜述[J];計算機科學;2011年04期
相關(guān)博士學位論文 前1條
1 唐滔;面向CPU-GPU異構(gòu)并行系統(tǒng)的編程模型與編譯優(yōu)化關(guān)鍵技術(shù)研究[D];國防科學技術(shù)大學;2012年
相關(guān)碩士學位論文 前1條
1 劉丹丹;面向異構(gòu)多核處理器的統(tǒng)一編程及分開編譯設(shè)計與實現(xiàn)[D];中國科學技術(shù)大學;2015年
,本文編號:1641518
本文鏈接:http://sikaile.net/kejilunwen/jisuanjikexuelunwen/1641518.html