【摘要】：隨著通信網絡的迅猛發(fā)展,用戶數量以及用戶對業(yè)務的流量需求也急劇增加,但網絡部署復雜、網絡架構垂直且封閉以及電信業(yè)務總量與收入不匹配等諸多問題使得當今的網絡面臨著巨大挑戰(zhàn)。網絡功能虛擬化(Network Function Virtualization,NFV)技術的出現和發(fā)展極大推進了通信網絡架構的演進,在基于NFV技術的網絡中,網絡功能將不再依賴于專用的硬件設備,而是通過軟件技術加以實現,從而部署在商用的硬件平臺上。服務功能鏈(Service Function Chain,SFC)的應用作為NFV技術的一個重要用途,能夠靈活地管理特定服務以及應用的流量,并且SFC支持網絡中細粒度和彈性的服務交付、服務功能的修改以及負載的移動,因此受到了國內外研究界的廣泛關注。本文關注點就是在引入NFV技術的網絡中的虛擬服務功能鏈的部署和遷移問題。在NFV網絡中,因物理資源本身的限制而可能產生資源碎片,造成網絡中資源總量足夠,但每臺主機上資源不足而無法部署服務功能鏈,降低網絡中可部署的服務功能鏈的數目。因此針對上述問題,我們以最大化資源受限的網絡中服務的SFCs的數目為目標,在滿足網絡服務請求端到端時延要求的前提下,建立了數學模型,提出基于貪心算法的虛擬服務功能鏈的部署方案并進行仿真驗證,結果表明該部署方案能夠有效的保證最大化資源受限的網絡中服務的SFCs的數目,在充分的利用網絡中物理主機和物理交換機的資源的同時也達到了兼顧網絡性能的目的。接下來,出于用戶的移動性以及網絡中負載均衡的考慮,實現資源的有效分配和網絡的靈活性,研究虛擬服務功能鏈的遷移工作是十分必要的,但目前關于虛擬SFC的遷移的工作并不多。本文中特別考慮了實際網絡功能的特性,針對SFCs的遷移重配置問題,將網絡性能和網絡中資源的合理利用這兩個重要因素納入考察,建立了數學模型,提出一種面向時延優(yōu)化的服務功能鏈重配置策略,即基于改進遺傳算法的SFCs遷移重配置方案,進行仿真驗證。通過與基于貪心的遷移重配置算法以及隨機遷移重配置算法等的對比,說明了基于改進遺傳算法的虛擬服務功能鏈遷移重配置方案在有效的保證了網絡性能的同時兼顧網絡物理資源。本文對NFV技術以及服務功能鏈的總結分析,設計了NFV網絡中虛擬服務功能鏈的部署和遷移重配置方案,為NFV網絡中服務功能鏈的部署和遷移問題提供了新的思路。
[Abstract]:With the rapid development of communication network, the number of users and the traffic demand of users have increased rapidly, but the network deployment is complex. Many problems, such as vertical and closed network architecture and mismatch between the total amount of telecom services and revenue, make today's network face great challenges. The emergence and development of network function virtualization (Network Function Virtualization,NFV) technology has greatly promoted the evolution of communication network architecture. In the network based on NFV technology, network function will no longer depend on special hardware devices, but will be realized through software technology. It is then deployed on commercial hardware platforms. As an important use of NFV technology, service function chain (Service Function Chain,SFC) can flexibly manage specific services and application traffic, and SFC supports fine-grained and flexible service delivery in a network. The modification of service function and the movement of load have attracted wide attention from domestic and foreign researchers. This paper focuses on the deployment and migration of virtual service function chain in the network with NFV technology. In NFV network, due to the limitation of physical resources, the resource fragmentation may occur, which results in the sufficient amount of resources in the network, but the insufficient resources on each host can not deploy the service functional chain, thus reducing the number of deployable service functional chains in the network. Therefore, in order to maximize the number of SFCs services in resource-constrained networks, we establish a mathematical model on the premise of satisfying the end-to-end delay requirements of network service requests. A deployment scheme of virtual service functional chain based on greedy algorithm is proposed and verified by simulation. The results show that the proposed deployment scheme can effectively guarantee the maximum number of SFCs of services in resource-constrained networks. The resource of the physical host and the physical switch in the network is fully utilized, and the network performance is also taken into account at the same time. Then, considering the mobility of users and the load balance in the network, it is necessary to study the migration of virtual service function chain to realize the effective allocation of resources and the flexibility of the network. However, there is not much work on the migration of virtual SFC. In this paper, the characteristics of the actual network function are specially considered. Aiming at the problem of migration and reconfiguration of SFCs, two important factors, network performance and rational utilization of resources in the network, are taken into account, and the mathematical model is established. A delay-oriented reconfiguration strategy for service function chain is proposed, that is, the SFCs migration reconfiguration scheme based on improved genetic algorithm, which is verified by simulation. Compared with greedy transfer reconfiguration algorithm and random migration reconfiguration algorithm, It is shown that the virtual service function chain migration and reconfiguration scheme based on improved genetic algorithm can effectively guarantee the network performance and take into account the network physical resources at the same time. In this paper, the NFV technology and service function chain are summarized and analyzed, and the deployment and migration reconfiguration scheme of virtual service function chain in NFV network is designed, which provides a new idea for the deployment and migration of service function chain in NFV network.
【學位授予單位】：電子科技大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN915.0

【參考文獻】

相關期刊論文 前7條

1 鄭舒;王曉東;;基于NFV架構的IMS核心網實現方案[J];電信工程技術與標準化;2016年05期

2 蔣銘;于益俊;;NFV與VNF架構演進與網絡架構未來[J];信息通信技術;2016年01期

3 劉鎏;虞紅芳;;基于資源拆分的虛擬網絡功能服務鏈映射算法[J];計算機應用研究;2016年08期

4 薛海強;張昊;;網絡功能虛擬化及其標準化[J];中興通訊技術;2015年02期

5 王殿君;;基于改進A~*算法的室內移動機器人路徑規(guī)劃[J];清華大學學報(自然科學版);2012年08期

6 劉鯖潔;陳桂明;劉小方;;基于矩陣編碼的遺傳算法研究[J];計算機工程;2011年13期

7 余有明;劉玉樹;閻光偉;;遺傳算法的編碼理論與應用[J];計算機工程與應用;2006年03期

相關碩士學位論文 前1條

1 呂晨;網絡服務功能鏈拓撲設計及映射研究[D];電子科技大學;2016年

，

本文編號：2221298