本文選題：領(lǐng)域應(yīng)用 + 可重構(gòu)計(jì)算�。� 參考：《解放軍信息工程大學(xué)》2015年博士論文

【摘要】：當(dāng)前科學(xué)計(jì)算領(lǐng)域的處理部件大致可分為通用計(jì)算器件和專(zhuān)用計(jì)算器件。通用計(jì)算器件具有通用性強(qiáng)、但效能不高的特點(diǎn),專(zhuān)用計(jì)算器件則具有靈活性差但效能高的特點(diǎn)。可重構(gòu)計(jì)算器件的興起成為這兩種計(jì)算模式的緩沖帶,不僅能夠通過(guò)重構(gòu)成為完成某項(xiàng)功能的高效專(zhuān)用器件,而且可以通過(guò)動(dòng)態(tài)重構(gòu)靈活地實(shí)現(xiàn)不同功能的切換。而目前通用的可重構(gòu)計(jì)算,為了滿(mǎn)足通用性、靈活性的要求,其重構(gòu)粒度相對(duì)較細(xì),這樣就使得對(duì)其管理、調(diào)度、重構(gòu)等代價(jià)相對(duì)較大。因此,在豐富和發(fā)展了可重構(gòu)理論與技術(shù)的基礎(chǔ)上,PRCA(Proactive Reconfigurable Computing Architecture)體系結(jié)構(gòu)理念,其引入構(gòu)件概念對(duì)重構(gòu)粒子進(jìn)行擴(kuò)展。因此,從邏輯門(mén)、片內(nèi)邏輯塊、IP核、部件直到子系統(tǒng)都是多維重構(gòu)的“可重構(gòu)粒子”。在PRCA系統(tǒng)中,基于對(duì)“應(yīng)用的屬性、服務(wù)質(zhì)量、負(fù)載變化、系統(tǒng)構(gòu)件資源”的認(rèn)知,來(lái)識(shí)別應(yīng)用的需求、應(yīng)用的變化,以及系統(tǒng)中可以利用的處理資源。依據(jù)盡可能高效、或可信安全的原則,在應(yīng)用需求與計(jì)算資源間動(dòng)態(tài)決策出效能最優(yōu)的匹配,構(gòu)建出適合于應(yīng)用需求的處理結(jié)構(gòu)。并且該結(jié)構(gòu)隨著應(yīng)用的變化,如：計(jì)算進(jìn)展階段、處理負(fù)荷等的變化,而進(jìn)行結(jié)構(gòu)的主動(dòng)變更。這樣,就實(shí)現(xiàn)了結(jié)構(gòu)適應(yīng)應(yīng)用的理念。PRCA體系結(jié)構(gòu)具有很多優(yōu)點(diǎn),當(dāng)然也存在相當(dāng)多的挑戰(zhàn)�；诖吮尘�,本文主要解決其在實(shí)現(xiàn)過(guò)程中的幾個(gè)關(guān)鍵問(wèn)題,本文的內(nèi)容也圍繞著這幾個(gè)問(wèn)題展開(kāi),主要工作及創(chuàng)新點(diǎn)包括：1.研究了基于領(lǐng)域應(yīng)用的特征分析方法,提出了多維度應(yīng)用特征模型。對(duì)于領(lǐng)域內(nèi)有限應(yīng)用的集合,存在著最適宜的處理結(jié)構(gòu),在滿(mǎn)足領(lǐng)域內(nèi)應(yīng)用的基礎(chǔ)上達(dá)到效能最優(yōu)。而特征是領(lǐng)域應(yīng)用本質(zhì)屬性的反映,因此基于對(duì)領(lǐng)域應(yīng)用特征的分析,可以建立最合適的計(jì)算模型,為使用和構(gòu)建最合適的處理部件提供必要支撐。特征模型的方法已廣泛應(yīng)用于多個(gè)領(lǐng)域,如制造業(yè)、自然語(yǔ)言分析、軟件設(shè)計(jì)等。本文把特征模型的方法引入到結(jié)構(gòu)設(shè)計(jì)當(dāng)中,建立了以特征向量、特征矩陣模型的應(yīng)用特征表達(dá)形式,給出了特征相關(guān)概念的定義,研究了特征的運(yùn)算方法和相關(guān)性質(zhì),并建立了從特征樹(shù)建立、特征量化、特征空間建立到特征規(guī)約聚類(lèi)的領(lǐng)域應(yīng)用算法建模過(guò)程。2.研究了基于算核的應(yīng)用算法實(shí)現(xiàn)方法,提出了算核利用率均等(EUCI)的算核求解算法。基本功能單元設(shè)計(jì)的好壞,很大程度上決定了體系結(jié)構(gòu)的計(jì)算效率。因此,本文基于算法特征,把計(jì)算粒化分解為一個(gè)個(gè)相對(duì)獨(dú)立的子過(guò)程,使領(lǐng)域應(yīng)用劃分為具有共性的算核,為提高硬件資源計(jì)算的復(fù)用程度,使其具有高性能和靈活性,提供了理論基礎(chǔ)。把效能控制轉(zhuǎn)化為資源利用率問(wèn)題,進(jìn)而表達(dá)為算核利用率問(wèn)題,提出了算核利用率均等迭代算法來(lái)求解算核,達(dá)到了高效能的目的。3.研究了可重構(gòu)計(jì)算模型,提出了基于元結(jié)構(gòu)的計(jì)算模型。計(jì)算模型是算法設(shè)計(jì)者與體系結(jié)構(gòu)之間的一個(gè)橋梁,是算法設(shè)計(jì)和分析的基礎(chǔ)。原有的時(shí)間、空間二維計(jì)算模型受制于主頻、功率等因素,即能耗因素。因此引入功耗,作為獨(dú)立于空間和時(shí)間的一維約束,使結(jié)構(gòu)受解算目標(biāo)的應(yīng)用驅(qū)動(dòng),兼顧高效計(jì)算和靈活性。并且,為了屏蔽異構(gòu)處理資源的差異,將硬件結(jié)構(gòu)劃分為應(yīng)用結(jié)構(gòu)和元結(jié)構(gòu),并論述了元結(jié)構(gòu)的組合和轉(zhuǎn)化形式。為了實(shí)現(xiàn)效能計(jì)算,在時(shí)、空計(jì)算的基礎(chǔ)上,提出了基于元結(jié)構(gòu)的時(shí)空能計(jì)算模型,分析了模型的時(shí)間、空間、能耗不同維度特性。建立了異構(gòu)資源的統(tǒng)一結(jié)構(gòu)視圖。4.研究了從應(yīng)用到計(jì)算結(jié)構(gòu)的映射方法,提出了基于模糊化效用函數(shù)的二人非零和博弈算法。對(duì)于具體計(jì)算的子任務(wù)要分配到由不同類(lèi)型的處理器構(gòu)成的異構(gòu)計(jì)算系統(tǒng)中,有多種對(duì)應(yīng)方法。不同的映射結(jié)果性能不同,甚至差距很大。本文從模糊化代價(jià)函數(shù)入手,以既競(jìng)爭(zhēng)又合作的復(fù)雜關(guān)系為處理對(duì)象,采用博弈論方法,實(shí)現(xiàn)算法和結(jié)構(gòu)的多對(duì)多映射,提供了一個(gè)能平衡任務(wù)和資源均衡收益的任務(wù)到計(jì)算結(jié)構(gòu)的映射算法。實(shí)驗(yàn)表明,該方法針對(duì)具體應(yīng)用、環(huán)境、運(yùn)行狀況等任務(wù)和系統(tǒng)資源參數(shù),達(dá)到了平衡應(yīng)用和資源的各種效益,實(shí)現(xiàn)了應(yīng)用和系統(tǒng)的均衡優(yōu)化目標(biāo)。5.從不同測(cè)度空間研究了體系結(jié)構(gòu)的評(píng)價(jià)方法,提出了多維聯(lián)合評(píng)價(jià)體系。為了評(píng)價(jià)多維的計(jì)算模型,建立了多維的評(píng)價(jià)體系,研究了不同維度間評(píng)價(jià)指標(biāo)的相互關(guān)系。并以多維評(píng)價(jià)體系為目標(biāo),構(gòu)建了面向領(lǐng)域應(yīng)用可重構(gòu)計(jì)算的驗(yàn)證系統(tǒng)和驗(yàn)證平臺(tái),綜合分析了測(cè)試結(jié)果,論證了面向應(yīng)用特征的計(jì)算方式的正確性和高效能。
[Abstract]:At present, the processing parts in the field of scientific computing can be divided into general calculator parts and special computing devices. The general calculator has the characteristics of strong generality, but its efficiency is not high. The special calculator has the characteristics of poor flexibility but high efficiency. The rise of reconfigurable computing devices is the buffer zone of these two computing modes. By refactoring, it is an efficient and special device to complete a certain function, and can be flexible to switch different functions through dynamic reconfiguration. At present, the reconfigurable computing is relatively fine in order to meet the generality and flexibility, so that the cost of management, scheduling and reconfiguration is relatively large. On the basis of enriching and developing Reconfigurable Theory and technology, the concept of PRCA (Proactive Reconfigurable Computing Architecture) architecture is introduced into the component concept to expand the reconfigurable particles. Therefore, from logical gates, intra chip logic blocks, IP cores, components to subsystems are "reconfigurable particles" of multidimensional reconfiguration. In PRCA systems, the system is a "reconfigurable particle". Based on the cognition of "application attributes, quality of service, load change, system component resources", the application needs, application changes, and the processing resources that can be used in the system are identified. Based on the principle of as high as possible as possible, or trusted security, the optimal matching between application requirements and computing resources is constructed. The structure is suitable for application requirements. And with the change of application, such as the change of computing progress, the change of load and so on, the structure is changed actively. So, the idea.PRCA architecture that realizes the application of structure has many advantages, and of course there are quite a lot of challenges. Based on this background, this paper mainly In order to solve several key problems in the process of implementation, the content of this paper also revolves around these problems. The main work and innovation points include: 1. research the feature analysis method based on the domain application, and put forward the multi-dimensional application feature model. Based on the application of the domain, the performance is optimal. The feature is the reflection of the essential properties of the domain application. Therefore, based on the analysis of the domain application features, the most suitable computing model can be established to provide the necessary support for the use and construction of the most suitable processing components. The method of feature model has been widely used in many fields, such as manufacturing, In this paper, the method of feature model is introduced into the structure design, the feature vector and feature matrix model are established, the definition of feature correlation concept is given, the operation method and correlation quality are studied, and the feature tree is established, the feature is quantified and the feature is set up. The application algorithm modeling process of space based on feature protocol clustering.2. studies the implementation method based on kernel based application algorithm, and proposes a kernel solving algorithm for kernel utilization ratio (EUCI). The design of basic functional units determines the computational efficiency of the architecture to a large extent. Therefore, this paper is based on the algorithm features. The calculation of granular decomposition into a relatively independent subprocess makes the domain application divided into a common kernel. In order to improve the reuse degree of the hardware resource calculation and make it have high performance and flexibility, it provides the theoretical basis. The efficiency control is transformed into the resource utilization rate problem, and then it is expressed as the kernel utilization rate problem, and the calculation kernel is put forward. The utilization ratio iterative algorithm is used to solve the kernel. The reconfigurable computing model is studied by.3.. The calculation model based on the meta structure is proposed. The calculation model is a bridge between the algorithm designer and the architecture. It is the basis of the algorithm design and analysis. The original time, the spatial two-dimensional calculation model is subject to the master. Frequency, power and other factors are the factors of energy consumption. Therefore, power consumption is introduced as one dimension constraint independent of space and time, which makes the structure be driven by the application of the target and takes account of the high efficiency calculation and flexibility. In order to shield the difference of heterogeneous processing resources, the hardware structure is divided into application structure and meta structure, and the combination of the meta structure is discussed. In order to realize efficiency calculation, on the basis of time and space computing, a spatio-temporal energy calculation model based on meta structure is proposed, and the time, space and energy consumption of the model are analyzed. A unified structure view.4. for heterogeneous resources is set up to study the mapping method from application to computing structure, and the fuzzy efficiency is proposed. The two person non zero sum game algorithm of function is used. There are a variety of corresponding methods for the subtasks of specific computation to be assigned to heterogeneous computing systems composed of different types of processors. The performance of different mapping results is different, and the difference is very large. This paper starts with the fuzzy cost function and deals with the complex relationship of both competition and cooperation. By using the method of game theory, the multi to multi mapping of algorithms and structures is realized, and a mapping algorithm which can balance task and resource equilibrium income to computing structure is provided. The experiment shows that the method has achieved various benefits in balancing application and resources for specific applications, environment, operation status and other tasks and system resource parameters. In order to evaluate the multidimensional model, a multi-dimensional evaluation system is set up, and the interrelationship between the evaluation indexes of different dimensions is studied, and the multi-dimensional evaluation system is set up as the goal, and the.5. is constructed. A verification system and a verification platform for domain application reconfigurable computing are presented. The test results are analyzed synthetically, and the correctness and efficiency of the application oriented computing method are demonstrated.
【學(xué)位授予單位】：解放軍信息工程大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：TP18

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 潘大勝;;分布式異構(gòu)資源庫(kù)共享機(jī)制研究[J];吉林師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年04期

2 吳金龍;;云計(jì)算異構(gòu)資源整合的分析與應(yīng)用[J];電力與能源;2012年06期

3 孟凡靜;李國(guó)朋;;分布式異構(gòu)數(shù)字資源整合概述[J];中華醫(yī)學(xué)圖書(shū)情報(bào)雜志;2006年05期

4 汪洋;集合異構(gòu)資源實(shí)現(xiàn)集中化管理[J];中國(guó)計(jì)算機(jī)用戶(hù);2004年35期

5 王立;王倩;馬朝東;;基于池化技術(shù)的資源復(fù)制方案的設(shè)計(jì)與實(shí)現(xiàn)[J];計(jì)算機(jī)科學(xué);2012年S3期

6 吳姍姍;;面向信息柵格的異構(gòu)資源通用管理機(jī)制[J];指揮信息系統(tǒng)與技術(shù);2012年02期

7 辛波;鄭高;正九;王嵩;;云環(huán)境中基于異構(gòu)資源的資源分配算法[J];電子技術(shù);2014年06期

8 黃偉紅,張福炎;EduMedia:教學(xué)數(shù)字視頻圖書(shū)館中異構(gòu)資源的統(tǒng)一描述[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2001年04期

9 李國(guó)棟;李樨;柳長(zhǎng)安;;基于Web Services異構(gòu)資源匯集與共享系統(tǒng)[J];計(jì)算機(jī)技術(shù)與發(fā)展;2008年12期

10 桂林斌;;基于J2EE的異構(gòu)資源統(tǒng)一檢索系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J];計(jì)算機(jī)與數(shù)字工程;2009年04期

相關(guān)重要報(bào)紙文章 前1條

1 IBM大數(shù)據(jù)專(zhuān)家 James Kobielus　范范 編譯;大數(shù)據(jù)需要數(shù)據(jù)虛擬化[N];網(wǎng)絡(luò)世界;2013年

相關(guān)博士學(xué)位論文 前3條

1 黃輝;分布式異構(gòu)模型組合與執(zhí)行問(wèn)題研究[D];華中科技大學(xué);2016年

2 李柏楠;面向領(lǐng)域應(yīng)用的可重構(gòu)系統(tǒng)關(guān)鍵技術(shù)研究[D];解放軍信息工程大學(xué);2015年

3 俞時(shí);異構(gòu)資源中基于本體的信息互操作性研究[D];東華大學(xué);2003年

相關(guān)碩士學(xué)位論文 前2條

1 陳硒;異構(gòu)服務(wù)組裝平臺(tái)中執(zhí)行引擎的設(shè)計(jì)與實(shí)現(xiàn)[D];中南大學(xué);2007年

2 賀云仙;基于本體的初中教學(xué)資源共享技術(shù)研究[D];南京師范大學(xué);2014年

，

本文編號(hào)：2017941