Distributed Approach to Adaptive SDN Controller Placement Problem

碩士 === 國立交通大學 === 網路工程研究所 === 107 === Software defined network (SDN) has emerged as a new paradigm that provides network programmability, flexibility and efficient network management. The main idea of SDN is splitting the network into application layer, control layer and data layer. In this splittin...

Full description

Bibliographic Details
Main Authors: Liu, Wei-Li, 劉威利
Other Authors: Yen, Li-Hsing
Format: Others
Language:en_US
Published: 2019
Online Access:http://ndltd.ncl.edu.tw/handle/3e5waf
id ndltd-TW-107NCTU5726043
record_format oai_dc
spelling ndltd-TW-107NCTU57260432019-11-26T05:16:53Z http://ndltd.ncl.edu.tw/handle/3e5waf Distributed Approach to Adaptive SDN Controller Placement Problem 分散式適應方法處理SDN 網路控制器佈署問題 Liu, Wei-Li 劉威利 碩士 國立交通大學 網路工程研究所 107 Software defined network (SDN) has emerged as a new paradigm that provides network programmability, flexibility and efficient network management. The main idea of SDN is splitting the network into application layer, control layer and data layer. In this splitting concept, controller is the brain of SDN network, which manages SDN switch. SDN network performance will be affected by controller location and some metrics such as controller-switch latency, inter-controller latency and controller capacity. It is not easy to determine the location for controllers and number of controllers requirement while meeting the constraint. There existing centralized solution that increases cost and not satisfies constraint when the network dynamics. In this paper, we proposed a potential game which is a distributed mechanism to solve controller placement problem (CPP). Meanwhile, our distributed mechanism have self-stabilizing property that it can adaptively change solution when the network dynamics such as controller failure, link failure and adding nodes. Also, we present our game have exact potential function and end up to Nash equilibrium (NE) eventually. Simulation result shows our approach have adaptability property Yen, Li-Hsing 嚴力行 2019 學位論文 ; thesis 38 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立交通大學 === 網路工程研究所 === 107 === Software defined network (SDN) has emerged as a new paradigm that provides network programmability, flexibility and efficient network management. The main idea of SDN is splitting the network into application layer, control layer and data layer. In this splitting concept, controller is the brain of SDN network, which manages SDN switch. SDN network performance will be affected by controller location and some metrics such as controller-switch latency, inter-controller latency and controller capacity. It is not easy to determine the location for controllers and number of controllers requirement while meeting the constraint. There existing centralized solution that increases cost and not satisfies constraint when the network dynamics. In this paper, we proposed a potential game which is a distributed mechanism to solve controller placement problem (CPP). Meanwhile, our distributed mechanism have self-stabilizing property that it can adaptively change solution when the network dynamics such as controller failure, link failure and adding nodes. Also, we present our game have exact potential function and end up to Nash equilibrium (NE) eventually. Simulation result shows our approach have adaptability property
author2 Yen, Li-Hsing
author_facet Yen, Li-Hsing
Liu, Wei-Li
劉威利
author Liu, Wei-Li
劉威利
spellingShingle Liu, Wei-Li
劉威利
Distributed Approach to Adaptive SDN Controller Placement Problem
author_sort Liu, Wei-Li
title Distributed Approach to Adaptive SDN Controller Placement Problem
title_short Distributed Approach to Adaptive SDN Controller Placement Problem
title_full Distributed Approach to Adaptive SDN Controller Placement Problem
title_fullStr Distributed Approach to Adaptive SDN Controller Placement Problem
title_full_unstemmed Distributed Approach to Adaptive SDN Controller Placement Problem
title_sort distributed approach to adaptive sdn controller placement problem
publishDate 2019
url http://ndltd.ncl.edu.tw/handle/3e5waf
work_keys_str_mv AT liuweili distributedapproachtoadaptivesdncontrollerplacementproblem
AT liúwēilì distributedapproachtoadaptivesdncontrollerplacementproblem
AT liuweili fēnsànshìshìyīngfāngfǎchùlǐsdnwǎnglùkòngzhìqìbùshǔwèntí
AT liúwēilì fēnsànshìshìyīngfāngfǎchùlǐsdnwǎnglùkòngzhìqìbùshǔwèntí
_version_ 1719296566939877376