WEKO3
アイテム
高信頼OpenFlowネットワークのための品質劣化リンクのネットワーク協調形監視・同定に関する研究
https://doi.org/10.18997/00008350
https://doi.org/10.18997/00008350d77aa946-950a-48d6-a691-3f2f131d8c8e
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
jou_k_354.pdf (1.6 MB)
|
|
| アイテムタイプ | 学位論文 = Thesis or Dissertation(1) | |||||||
|---|---|---|---|---|---|---|---|---|
| 公開日 | 2021-06-09 | |||||||
| 資源タイプ | ||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_db06 | |||||||
| 資源タイプ | doctoral thesis | |||||||
| タイトル | ||||||||
| タイトル | Studies on Network-Assisted Monitoring and Location of Performance Degenerate Links for Resilient OpenFlow Networks | |||||||
| 言語 | en | |||||||
| タイトル | ||||||||
| タイトル | 高信頼OpenFlowネットワークのための品質劣化リンクのネットワーク協調形監視・同定に関する研究 | |||||||
| 言語 | ja | |||||||
| 言語 | ||||||||
| 言語 | eng | |||||||
| 著者 |
Nguyen, Minh Tri
× Nguyen, Minh Tri
|
|||||||
| 抄録 | ||||||||
| 内容記述タイプ | Abstract | |||||||
| 内容記述 | As practical realization of Software Defined Networking （SDN） technology, the OpenFlow-based networks are becoming widespread as a replacement solution for traditional network not only in data centers but also in enterprise networks and wide area networks or so-called SD-WAN. By decoupling the control-plane and data-plane, SDN in general and OpenFlow in particular become more flexible and simpler to design, configure, operate, and monitor. On the control-plane, centralized controllers deploy services and features in the network by programming. On a data-plane, switches forward packets based on rules from controllers. The prevalence of cloud computing and contents delivery networking has led to demand for OpenFlow-based centrally-managed networks with dynamic and flexible traffic engineering. To maintain a high level of network service quality, it is essential for network operators to detect and locate degenerate links. This research presents a network-assisted measurement framework for resilient OpenFlow networks to monitor all links in both directions to promptly and efficiently detect and locate degenerate links, with minimizing the load on both the data-plane and control-plane incurred by measurement. In contrast to existing works, our framework combines an active measurement by probing multicast packets from a measurement host to switches in the network along a designed route and a passive measurement by collecting the statistic information of probing flows （flow-stats） at required switch ports in an appropriate sequential order to access switches that is determined dynamically. We propose two kinds of route schemes for probe packets. The first one is the shortest path tree-based route schemes including a route with minimum length paths, called Model 1, and a variant from it, Model 2, with a fewer number of measurement paths. Both models use Dijkstra’s shortest path algorithm first to build a tree on an undirected graph from the measurement host （MH）. These routes keep probing paths as short as possible to overcome the loss of probe packets along the long route. However, their complex route tree needs more loads on the control-plane （the number of accesses to switches until locating all degenerate links）. Therefore, as a second route scheme, we propose a completely new route scheme, called the backbone-and-branch tree （BBT）, that can intentionally control the number of probing paths and their lengths. After creating a route for probe packets, by collecting the number of successful probe packets of the flow-stats, we can calculate the packet loss rate of a link or a segment of links and then locate high-loss links. An appropriate sequential order of accesses is proposed in this research to minimize the loads for the location process. In addition, an estimation of packet delay variance on each link （note both up and down directions of each full-duplex link are distinguished） in an OpenFlow network is presented in the dissertation. A notable feature is that packet delay variance is estimated based on monitoring arrival time intervals of probe packets instead of directly measuring packet delay time over a link. All evaluated results are done on a large-scale network topology based on the real network by numerical simulation to demonstrate the effectiveness of our proposals. | |||||||
| 目次 | ||||||||
| 内容記述タイプ | TableOfContents | |||||||
| 内容記述 | 1 Introduction||2 Network-Assisted Monitoring Framework||3 Effective Backbone-and-Branch Tree Route Scheme||4 Locating High Delay Variance Links by Packet Arrival Intervals||5 Discussions||6 Conclusions and Future Work | |||||||
| 備考 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | 九州工業大学博士学位論文 学位記番号：情工博甲第354号 学位授与年月日：令和3年3月25日 | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | active measurement | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | multicast probing | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | flow statistics | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | OpenFlow | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | SDN | |||||||
| アドバイザー | ||||||||
| 鶴, 正人 | ||||||||
| 学位授与番号 | ||||||||
| 学位授与番号 | 甲第354号 | |||||||
| 学位名 | ||||||||
| 学位名 | 博士（情報工学） | |||||||
| 学位授与年月日 | ||||||||
| 学位授与年月日 | 2021-03-25 | |||||||
| 学位授与機関 | ||||||||
| 学位授与機関識別子Scheme | kakenhi | |||||||
| 学位授与機関識別子 | 17104 | |||||||
| 学位授与機関名 | 九州工業大学 | |||||||
| 学位授与年度 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | 令和2年度 | |||||||
| 出版タイプ | ||||||||
| 出版タイプ | VoR | |||||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||
| アクセス権 | ||||||||
| アクセス権 | open access | |||||||
| アクセス権URI | http://purl.org/coar/access_right/c_abf2 | |||||||
| ID登録 | ||||||||
| ID登録 | 10.18997/00008350 | |||||||
| ID登録タイプ | JaLC | |||||||
