本文選題：高效能計(jì)算 + 功耗管理�。� 參考：《國防科學(xué)技術(shù)大學(xué)》2012年博士論文

【摘要】：為滿足大型應(yīng)用不斷增長的計(jì)算性能需求，高端計(jì)算系統(tǒng)的規(guī)模越來越大，結(jié)構(gòu)越來越復(fù)雜，計(jì)算密度越來越高。功耗大、管理復(fù)雜、可靠性低、成本高等問題嚴(yán)重制約著大規(guī)模計(jì)算技術(shù)的進(jìn)一步發(fā)展。高端計(jì)算已經(jīng)由一味地追求高性能轉(zhuǎn)向綜合考慮系統(tǒng)產(chǎn)出率的高效能，力求在提高系統(tǒng)性能、魯棒性、易用性的同時(shí)，降低系統(tǒng)成本。 系統(tǒng)軟件是實(shí)現(xiàn)大規(guī)模系統(tǒng)高效能計(jì)算的關(guān)鍵環(huán)節(jié)。本文面向高效能計(jì)算，立足系統(tǒng)軟件，從功耗管理和用戶環(huán)境兩個(gè)方面展開研究，取得如下創(chuàng)新性成果： 1、為保障大規(guī)模計(jì)算系統(tǒng)在供電預(yù)算約束下的用電安全，提出了基于結(jié)點(diǎn)分類的系統(tǒng)峰值功耗管理模型PCNC及相應(yīng)的功耗控制算法，以可接受的管理開銷和系統(tǒng)性能損失，有效地控制系統(tǒng)的運(yùn)行峰值功耗。其創(chuàng)新點(diǎn)有：（A）按功耗特性和作用，將系統(tǒng)中的結(jié)點(diǎn)分為統(tǒng)計(jì)源結(jié)點(diǎn)集合、特權(quán)結(jié)點(diǎn)集合、候選結(jié)點(diǎn)集合和目標(biāo)結(jié)點(diǎn)集合，降低系統(tǒng)功耗管理的采樣和控制規(guī)模；（B）采用兩級(jí)閾值設(shè)置，將系統(tǒng)功耗分為安全、警戒和危險(xiǎn)三個(gè)狀態(tài)，以作業(yè)為基本調(diào)節(jié)單位，對(duì)不同狀態(tài)采取不同程度、不同結(jié)點(diǎn)范圍的功耗控制措施；（C）設(shè)計(jì)并研究了基于狀態(tài)和基于變化的兩類目標(biāo)結(jié)點(diǎn)選擇策略。實(shí)驗(yàn)表明，該功耗控制系統(tǒng)在控制效果最大損失7.4%的代價(jià)下，將控制開銷降低了76.3%，適用于大規(guī)模計(jì)算系統(tǒng)，兩類目標(biāo)結(jié)點(diǎn)選擇策略在系統(tǒng)性能損失分別為1.4%和1.1%的代價(jià)下，超標(biāo)功耗累積效應(yīng)各自降低73%和66%，優(yōu)化效果明顯。 2、針對(duì)大規(guī)模系統(tǒng)中未運(yùn)行作業(yè)的空閑活躍結(jié)點(diǎn)產(chǎn)生的能耗浪費(fèi)，提出了大規(guī)模系統(tǒng)空閑結(jié)點(diǎn)的功耗管理模型ASDMIN以及對(duì)空閑結(jié)點(diǎn)休眠深度的自適應(yīng)管理算法，以較小的響應(yīng)速率損失，有效降低系統(tǒng)空閑能耗。其創(chuàng)新點(diǎn)有：（A）多級(jí)儲(chǔ)備結(jié)構(gòu)：以當(dāng)前結(jié)點(diǎn)支持多個(gè)休眠狀態(tài)的硬件機(jī)制為基礎(chǔ)，將空閑結(jié)點(diǎn)按所處功耗狀態(tài)劃分為不同休眠等級(jí)的結(jié)點(diǎn)儲(chǔ)備集合；（B）隱蔽式狀態(tài)遷移：資源分配時(shí)，首先從最高級(jí)儲(chǔ)備池選取結(jié)點(diǎn)，只有當(dāng)高級(jí)儲(chǔ)備池中的結(jié)點(diǎn)不足以滿足應(yīng)用需求時(shí)，才由低一級(jí)儲(chǔ)備池中的結(jié)點(diǎn)補(bǔ)足，空閑結(jié)點(diǎn)的功耗狀態(tài)根據(jù)應(yīng)用負(fù)載需求動(dòng)態(tài)升降級(jí)，狀態(tài)遷移在結(jié)點(diǎn)處于空閑備用時(shí)進(jìn)行，狀態(tài)遷移的時(shí)間開銷不影響系統(tǒng)響應(yīng)速率；（C）自適應(yīng)控制算法：綜合考慮能耗與系統(tǒng)響應(yīng)速率兩個(gè)相互沖突的因素，設(shè)計(jì)了基于ASDMIN模型的資源分配與回收算法、休眠結(jié)點(diǎn)在不同休眠狀態(tài)之間的動(dòng)態(tài)升降級(jí)算法以及儲(chǔ)備額閾值動(dòng)態(tài)自適應(yīng)校準(zhǔn)算法。實(shí)驗(yàn)表明，ASDMIN方法在作業(yè)平均響應(yīng)延遲率僅增加8.85%的代價(jià)下，系統(tǒng)空閑結(jié)點(diǎn)功耗降低84.12%，，系統(tǒng)空閑功效提高了82.71%，優(yōu)化效果顯著。 3、針對(duì)傳統(tǒng)的、用戶登錄共享的并行計(jì)算系統(tǒng)使用環(huán)境中，系統(tǒng)使用、管理及用戶數(shù)據(jù)安全所面臨的問題，提出了高性能虛擬域技術(shù)，在保證系統(tǒng)高性能的前提下，為用戶提供專用的虛擬化巨型機(jī)環(huán)境。創(chuàng)新點(diǎn)包括：（A）為滿足用戶的不同需求，設(shè)計(jì)了通用虛擬域和專用虛擬域兩種用戶環(huán)境，并采用本地與全局相結(jié)合的兩級(jí)文件部署模式，優(yōu)化文件訪問性能；（B）基于環(huán)境提取和文件訪問路徑動(dòng)態(tài)轉(zhuǎn)換機(jī)制，實(shí)現(xiàn)計(jì)算陣列高性能計(jì)算域的動(dòng)態(tài)構(gòu)建；（C）提出影子系統(tǒng)文件設(shè)置和文件訪問穿透鏈接技術(shù)，實(shí)現(xiàn)虛擬計(jì)算環(huán)境文件系統(tǒng)的安全隔離和雙分區(qū)單映像存儲(chǔ)。實(shí)驗(yàn)表明，該虛擬計(jì)算環(huán)境符合LSB和POSIX標(biāo)準(zhǔn)，前端服務(wù)陣列的性能損失小于3%，后端計(jì)算陣列的性能損失小于0.5%，滿足高效能計(jì)算的需求。 4、為保障用戶對(duì)計(jì)算資源的需求，同時(shí)防止用戶占用過多資源而導(dǎo)致系統(tǒng)故障，提出了多粒度自適應(yīng)服務(wù)質(zhì)量保障機(jī)制。主要?jiǎng)?chuàng)新點(diǎn)包括：（A）多粒度服務(wù)質(zhì)量控制機(jī)制：分別以進(jìn)程、進(jìn)程組、用戶作業(yè)、虛擬用戶環(huán)境等多種粒度進(jìn)行資源管理和使用控制，滿足虛擬化環(huán)境中不同運(yùn)行實(shí)體的服務(wù)質(zhì)量需求；（B）自適應(yīng)的資源分配機(jī)制：設(shè)計(jì)了雙閾值自適應(yīng)資源限制機(jī)制，根據(jù)用戶需求和系統(tǒng)資源狀態(tài)，自適應(yīng)調(diào)整用戶資源使用限額，彌補(bǔ)了用戶資源預(yù)約的不準(zhǔn)確性；（C）終止目標(biāo)選擇策略：提出并分析了多種終止目標(biāo)選擇策略，以便在用戶的資源分配請(qǐng)求無法被滿足的情況下，合理終止某些運(yùn)行實(shí)體，保障系統(tǒng)的高效運(yùn)行。實(shí)驗(yàn)表明，本文提出的服務(wù)質(zhì)量保障技術(shù)將系統(tǒng)產(chǎn)出率提高了17.14%，而對(duì)應(yīng)用性能的影響不超過0.65%，可以忽略不計(jì)。 5、由于虛擬機(jī)的“隔離”效應(yīng)，傳統(tǒng)功耗管理技術(shù)在虛擬化環(huán)境中不能直接操作硬件，針對(duì)這一難題，提出了虛擬機(jī)和物理主機(jī)兩級(jí)功耗管理模型以及相應(yīng)的功耗管理算法。主要?jiǎng)?chuàng)新有：（A）兩級(jí)功耗管理：分別在虛擬機(jī)內(nèi)和虛擬化實(shí)現(xiàn)層實(shí)施功耗管理；（B）虛擬機(jī)中的功耗管理設(shè)施與機(jī)制：在虛擬機(jī)中引入虛設(shè)備功耗行為的監(jiān)測(cè)與統(tǒng)計(jì)，設(shè)計(jì)了虛設(shè)備速率動(dòng)態(tài)調(diào)節(jié)、虛設(shè)備動(dòng)態(tài)休眠、以及虛擬機(jī)動(dòng)態(tài)休眠等多種虛擬化功耗管理機(jī)制，實(shí)現(xiàn)了虛擬功耗狀態(tài)到物理功耗狀態(tài)的疊加映射，向虛擬機(jī)內(nèi)用戶提供了虛設(shè)備和虛擬機(jī)的功耗管理接口。實(shí)驗(yàn)表明，虛擬化功耗管理機(jī)制與傳統(tǒng)物理功耗管理機(jī)制完全兼容，基于虛擬化功耗管理機(jī)制實(shí)現(xiàn)的虛擬機(jī)內(nèi)功耗管理方案將系統(tǒng)能效優(yōu)化了2.75%，而功耗管理虛擬化對(duì)應(yīng)用性能的影響不超過0.4%。
[Abstract]:In order to meet the increasing demand for computing performance of large scale applications, the scale of the high end computing system is increasing, the structure is more complex, the calculation density is getting higher, the power consumption, the management complexity, the low reliability and the high cost seriously restrict the further development of the large-scale computing technology. The high-end computing has been blindly pursuing the high performance. Considering the high efficiency of system output rate, we strive to reduce system cost while improving system performance, robustness and ease of use.
System software is the key link to achieve high efficiency in large scale systems. This paper focuses on the efficiency calculation, based on the system software, from two aspects of power management and user environment, and obtain the following innovative results:
1, in order to ensure the power consumption security of large scale computing systems under the constraints of power supply budget, a system peak power management model PCNC based on node classification and the corresponding power control algorithm are proposed, which can effectively control the peak power consumption of the system with acceptable management overhead and system performance loss. The innovation point is: (A) according to the power characteristics and the power characteristics of the system. The nodes in the system are divided into the set of statistical source nodes, the set of privileged nodes, the set of candidate nodes and the set of target nodes to reduce the sampling and control scale of the power management of the system. (B) the system is set up by the two level threshold, and the system power is divided into security, alert and dangerous three states, with the operation as the basic regulation unit and the different state mining. The power control measures of different degree and different node range are taken. (C) the two types of target node selection strategy based on state and change are designed and studied. The experiment shows that the control system reduces the control overhead by 76.3% at the cost of the maximum control loss of 7.4%, and is suitable for the large scale computing system and the two target nodes. When the system performance loss is 1.4% and 1.1% respectively, the cumulative effect of over standard power consumption decreases by 73% and 66% respectively, and the optimization effect is obvious.
2, in view of the waste of idle active nodes in large scale systems, a power management model ASDMIN and an adaptive management algorithm for idle node dormant depth of large scale systems are proposed, which can reduce the idle energy consumption of the system effectively with smaller response rate loss. The innovation point is: (A) multilevel storage. Structure: Based on the hardware mechanism of the current node supporting multiple dormancy States, the free nodes are divided into the node reserve sets of different dormancy levels according to their power state; (B) hidden state migration: when resource allocation, the node is first selected from the most advanced reserve pool, only when the nodes in the advanced reserve pool are not enough to satisfy. When the requirements are applied, the nodes in the lower level reserve pool are complemented. The power state of the idle nodes is dynamically raised and lowered according to the application load demand. The state migration is carried out in the idle spare time. The time overhead of the state migration does not affect the response rate of the system; (C) the adaptive control algorithm: the energy consumption and the response rate of the system are considered. Two conflicting factors are designed, which are based on the resource allocation and recovery algorithm based on ASDMIN model, the dynamic lifting level algorithm and the reserve threshold dynamic adaptive calibration algorithm between dormancy nodes in different dormancy states. The experiment shows that the ASDMIN method has the idle node of the system under the cost of only increasing the delay rate of the job average response. The power consumption is reduced by 84.12%, the idle efficiency of the system is increased by 82.71%, and the optimization effect is remarkable.
3, in view of the traditional, user logon and shared parallel computing system, using the environment, system use, management and user data security, presents a high performance virtual domain technology to provide a special virtual giant environment for users under the premise of guaranteeing the high performance of the system. The innovation points include: (A) to satisfy the user's different In demand, two user environments of general virtual domain and special virtual domain are designed, and the two level file deployment mode combining local and global is used to optimize file access performance. (B) based on dynamic transformation mechanism of environment extraction and file access path, dynamic construction of computational array high energy computing domain is realized; (C) a shadow system file is proposed. It is shown that the virtual computing environment conforms to the LSB and POSIX standards, the performance loss of the front end service array is less than 3%, and the performance loss of the back end computing array is less than 0.5%, which satisfies the requirement of high efficiency calculation.
4, in order to guarantee the user's demand for computing resources and prevent the user from taking over too many resources and causing system failure, a multi granularity adaptive quality assurance mechanism is proposed. The main innovation points include: (A) multi granularity service quality control mechanism: resources such as process, process group, user job, virtual user environment, etc., respectively. Management and use control to meet the service quality requirements of different running entities in the virtualization environment; (B) adaptive resource allocation mechanism: a dual threshold adaptive resource restriction mechanism is designed, which adaptively adjusts the user resource usage limit according to the user needs and system resource status, and makes up the inaccuracy of the user's resource reservation; (C) To terminate the target selection strategy, a variety of terminating target selection strategies are proposed and analyzed in order to reasonably terminate some running entities and ensure the efficient operation of the system under the circumstances that the user's resource allocation request cannot be met. The experiment shows that the quality of service technology proposed in this paper raises the system output rate by 17.14% and is applied to the application. The impact of performance is not more than 0.65%, which can be ignored.
5, due to the "isolation" effect of the virtual machine, the traditional power management technology can not operate the hardware directly in the virtualization environment. Aiming at this problem, the two level power management model of virtual machine and physical host and the corresponding power management algorithm are proposed. The main innovations are: (A) two level power management: in virtual machine and virtual reality respectively The current layer implements power management; (B) power management facilities and mechanisms in virtual machines: monitoring and statistics are introduced in virtual machines, virtual device rate dynamic regulation, virtual equipment dynamic dormancy, and virtual machine dynamic dormancy are designed to realize virtual power state to physical work. The power management interface of virtual machines and virtual machines is provided to users in virtual machines. Experiments show that the virtualized power management mechanism is fully compatible with the traditional physical power management mechanism. The virtual machine internal power consumption management scheme based on virtual power management mechanism optimizes the energy efficiency of the system by 2.75%, and the power management system is used. The effect of physical virtualization on application performance is not more than 0.4%.

【學(xué)位授予單位】：國防科學(xué)技術(shù)大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2012
【分類號(hào)】：TP38;TP311.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 袁偉,張?jiān)迫?孫家昶,李玉成;國產(chǎn)萬億次機(jī)群系統(tǒng)NPB性能測(cè)試分析[J];計(jì)算機(jī)研究與發(fā)展;2005年06期

2 劉勇鵬;盧凱;劉勇燕;武林平;陳娟;;高性能計(jì)算中處理器功耗特征的評(píng)測(cè)與分析[J];計(jì)算機(jī)工程與科學(xué);2009年11期

3 谷立靜;周伏秋;孟輝;;我國數(shù)據(jù)中心能耗及能效水平研究[J];中國能源;2010年11期

相關(guān)博士學(xué)位論文 前1條

1 盧宇彤;面向高效能計(jì)算的大規(guī)模資源管理技術(shù)研究與實(shí)現(xiàn)[D];國防科學(xué)技術(shù)大學(xué);2009年

本文編號(hào)：1878984