本文選題：高密度計(jì)算 + 片上網(wǎng)絡(luò)��； 參考：《合肥工業(yè)大學(xué)》2015年碩士論文

【摘要】：隨著科技的發(fā)展和生產(chǎn)生活的信息化,諸多領(lǐng)域?qū)μ幚砥鞯挠?jì)算能力提出了更高的訴求。在算法方面,人們提出了多種高效的算法來(lái)解決實(shí)際問(wèn)題,其中一部分的發(fā)展趨勢(shì)是計(jì)算密度急劇提高和支持并行計(jì)算。而在計(jì)算平臺(tái)方面,也演變出了各種經(jīng)過(guò)特別優(yōu)化的專(zhuān)用平臺(tái)用于滿足嚴(yán)苛的需求。基于片上網(wǎng)絡(luò)(NoC)架構(gòu)的異構(gòu)多核處理器是其中一個(gè)重要的分支。研究提高異構(gòu)多核NoC處理器的性能,是半導(dǎo)體行業(yè)一個(gè)重要課題。本文對(duì)上述問(wèn)題,對(duì)面向高密度計(jì)算的NoC平臺(tái)上多發(fā)射技術(shù)進(jìn)行了研究和實(shí)驗(yàn)。本文所做主要工作如下：一、針對(duì)單發(fā)射原型系統(tǒng)NoC平臺(tái)進(jìn)行多發(fā)射改進(jìn)分析,提出了基于可擴(kuò)展發(fā)射通道的主控制器設(shè)計(jì),管理多發(fā)射的動(dòng)態(tài)調(diào)度。針對(duì)多發(fā)射系統(tǒng)運(yùn)算簇資源使用特性,設(shè)計(jì)了用于運(yùn)算簇資源動(dòng)態(tài)分配的空閑簇隊(duì)列和相應(yīng)排序算法。針對(duì)計(jì)算子任務(wù)的動(dòng)態(tài)訪存特性設(shè)計(jì)了地址映射關(guān)系表,用于動(dòng)態(tài)映射虛擬地址/實(shí)地址。針對(duì)動(dòng)態(tài)分配運(yùn)算簇和實(shí)地址的特性,設(shè)計(jì)了配置信息填充模塊。針對(duì)多發(fā)射改進(jìn)控制特性,進(jìn)行了原型系統(tǒng)的DDR去通道化設(shè)計(jì)和狀態(tài)網(wǎng)絡(luò)功能增強(qiáng)設(shè)計(jì),用于適應(yīng)多發(fā)射改進(jìn)。針對(duì)瓶頸子任務(wù),提出了支持局部算法級(jí)亂序多發(fā)射的優(yōu)化設(shè)計(jì)。二、對(duì)本文設(shè)計(jì)的多發(fā)射主控制器建立了軟件功能模型,并加入了時(shí)間變量用于獲取系統(tǒng)運(yùn)行的性能信息。在實(shí)驗(yàn)準(zhǔn)備階段,完成對(duì)一種高密度算法的分析與任務(wù)映射。隨后,使用軟件模型加載高密度算法對(duì)系統(tǒng)進(jìn)行了多發(fā)射功能驗(yàn)證和性能提升驗(yàn)證,并對(duì)比了是否開(kāi)啟局部算法級(jí)亂序多發(fā)射功能對(duì)系統(tǒng)性能提升的影響。根據(jù)性能分析結(jié)果,提出了優(yōu)化系統(tǒng)運(yùn)算簇構(gòu)成和算法映射的方法,并經(jīng)軟件模型驗(yàn)證有效。最后,根據(jù)奇數(shù)個(gè)算法前/后端計(jì)算與偶數(shù)個(gè)的差距提出了解決辦法,經(jīng)軟件模型驗(yàn)證有效。三、對(duì)本文設(shè)計(jì)的多發(fā)射主控制器進(jìn)行了RTL級(jí)代碼編寫(xiě),并進(jìn)行了系統(tǒng)硬件集成。首先使用ModelSim仿真驗(yàn)證了子任務(wù)規(guī)模4096：1縮比下的功能正確性,隨后在FPGA開(kāi)發(fā)板上完成全規(guī)模實(shí)驗(yàn),驗(yàn)證了設(shè)計(jì)的可行性和軟件模型獲取性能信息的準(zhǔn)確性。本文改進(jìn)設(shè)計(jì)的多發(fā)射系統(tǒng)具有如下特點(diǎn)：(1)任務(wù)調(diào)度方式適用于高計(jì)算密度類(lèi)型的算法,多發(fā)射方式適用于在高密度算法中進(jìn)行多路并行計(jì)算。(2)支持各發(fā)射通道進(jìn)行不同類(lèi)型計(jì)算。(3)多發(fā)射對(duì)象為計(jì)算子任務(wù),采用動(dòng)態(tài)調(diào)度方式。(4)多發(fā)射通道共享系統(tǒng)資源,通道數(shù)量可根據(jù)需要進(jìn)行擴(kuò)展。(5)根據(jù)不同發(fā)射通道動(dòng)態(tài)分配最優(yōu)運(yùn)算簇資源。(6)精簡(jiǎn)配置信息長(zhǎng)度,動(dòng)態(tài)映射實(shí)地址/虛擬地址空間。(7)具有可選擇開(kāi)啟的局部亂序多發(fā)射優(yōu)化功能。
[Abstract]:With the development of science and technology and the production of life information, put forward higher demands on the processor computing power in many fields. In the algorithm, it proposes several efficient algorithms to solve practical problems, which is part of the development trend of computing density increases sharply and parallel computing. In computing platform, but also the evolution of after a variety of special platform specially optimized to meet stringent requirements. Based on network on chip (NoC) architecture of heterogeneous multi-core processor is one of the important branches of research. To improve the performance of heterogeneous multi-core NoC processor, is an important issue in the semiconductor industry. Based on the above problems, opposite to high density calculation the NoC platform on the launch technology for research and experiment. The main work of this paper is as follows: first, for the single launch prototype system of NoC platform for multi emission improvement analysis, put forward The design of the main controller can extend the transmission channel based on dynamic scheduling and management of multiple transmissions. For many launch system cluster resource usage characteristics for free cluster queue cluster dynamic resource allocation and the corresponding sorting algorithm is designed. According to the dynamic calculation of sub task memory address mapping table design characteristics, for dynamic mapping virtual address / real address. According to the dynamic allocation of cluster and real address characteristics, design the configuration information filling module. For many improved emission control characteristics, the prototype system of DDR to channel design and network function design enhancement, to adapt to the multi launch. Aiming at the bottle neck improvement task, put forward the optimum design support local algorithm level reordering multiple transmit. Two, the multi transmitting main controller to establish software function model, and add the time variable for acquiring system The performance of information operation. In preparation for the experiment, analysis and mapping of tasks to complete a high density algorithm. Then, using the software model loading high density algorithm of multiple transmit function verification and performance improvement verification and comparison of whether to open the local algorithm level reordering multiple transmit function of system performance improvement according to the results of the performance analysis, the method of optimizing the system cluster structure and mapping algorithm, and the software model is validated. Finally, according to the odd and even before the algorithm / back-end computing a gap between the proposed solving methods, the software model is validated. Three, the design of the main controller for multi emission the RTL code, and the system hardware integration. Firstly, using ModelSim simulation to verify the sub task size 4096:1 scale under the functional correctness, then in the FPGA development board finish Full scale experiments to verify the accuracy of the model and the feasibility of software design to obtain performance information. This improved design of the transmission system has the following characteristics: (1) the task scheduling method is suitable for high density type calculation algorithm, multi emission mode applied in the multi-channel parallel computing method is of high density (2) support. The transmission channel for different kinds of calculation. (3) for computing multiple transmit sub tasks, using dynamic scheduling methods. (4) multi transmitting channel resource sharing system, the number of channels can be extended according to requirement. (5) according to the different transmission channel dynamic optimal allocation of cluster resources. (6) Streamline configuration information length. Dynamic address mapping real / virtual address space. (7) can choose to open the local order optimization function of multi transmitting.

【學(xué)位授予單位】：合肥工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：TP332

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 曾錫山;宋美妍;張小波;;基于多Agent系統(tǒng)的協(xié)同任務(wù)求解算法研究[J];計(jì)算機(jī)應(yīng)用;2006年08期

2 侯亮;陳峰;溫志嘉;;跨企業(yè)產(chǎn)品協(xié)同開(kāi)發(fā)中的設(shè)計(jì)任務(wù)分解與分配[J];浙江大學(xué)學(xué)報(bào)(工學(xué)版);2007年12期

3 林金芳;張?zhí)靹?;信息聚焦下的任務(wù)分解方法[J];計(jì)算機(jī)系統(tǒng)應(yīng)用;2009年02期

4 胡偉,朱林,常明;基于分級(jí)規(guī)劃的CAPP工藝決策方法研究[J];機(jī)械與電子;2000年05期

5 彭志平;李紹平;;分層強(qiáng)化學(xué)習(xí)研究進(jìn)展[J];計(jì)算機(jī)應(yīng)用研究;2008年04期

6 劉乃若,董金祥,李善平,陳剛;ZD-PDM中面向目標(biāo)的項(xiàng)目管理模型的設(shè)計(jì)[J];計(jì)算機(jī)研究與發(fā)展;1999年10期

7 趙燁;;參數(shù)相關(guān)任務(wù)中Option算法的子任務(wù)重用[J];煤炭技術(shù);2014年01期

8 李崇;;引入式項(xiàng)目教學(xué)方法在高職院校計(jì)算機(jī)軟件教學(xué)中的應(yīng)用[J];計(jì)算機(jī)教育;2010年01期

9 錢(qián)大可,琚春華;MAS中平等個(gè)體的交互[J];微電子學(xué)與計(jì)算機(jī);2004年04期

10 陳大川;張榮國(guó);黃付亮;劉q;;PSO算法在子任務(wù)分配中的應(yīng)用[J];計(jì)算機(jī)工程;2011年24期

相關(guān)博士學(xué)位論文 前1條

1 沈晶;分層強(qiáng)化學(xué)習(xí)方法研究[D];哈爾濱工程大學(xué);2006年

相關(guān)碩士學(xué)位論文 前3條

1 吳凌云;面向高密度計(jì)算的NoC平臺(tái)多發(fā)射技術(shù)研究[D];合肥工業(yè)大學(xué);2015年

2 傅一峰;服務(wù)Agent的聯(lián)盟方法及其應(yīng)用[D];上海交通大學(xué);2010年

3 岳峰;面向PC的輕量級(jí)網(wǎng)格中間件的設(shè)計(jì)與實(shí)現(xiàn)[D];同濟(jì)大學(xué);2006年

，

本文編號(hào)：1762838