2020年1月
Dynamic optimization of multicast active probing path to locate lossy links for OpenFlow networks
International Conference on Information Networking
- ,
- ,
- 巻
- 2020-January
- 号
- 開始ページ
- 628
- 終了ページ
- 633
- 記述言語
- 英語
- 掲載種別
- 研究論文(国際会議プロシーディングス)
- DOI
- 10.1109/ICOIN48656.2020.9016438
© 2020 IEEE. To maintain a high quality of service in managed networks, detecting and locating high loss-rate links (i.e., lossy links that are likely congested or physically unstable) in a fast and efficient manner is required. In our previous work, we proposed a centrally-managed network-assisted framework of locating lossy links on OpenFlow networks. In the framework, the OpenFlow controller builds a multicast measurement route; a measurement host launches a series of multicast probe packets traversing all full-duplex links along the measurement route; and then the controller collects statistical information (flow-stats) on the arrival of those probe packets at different input ports on selected switches and compares them to narrow down and identify the locations of high loss-rate links. The number of accesses to switches in collecting the flow-stats until locating all lossy links should be as small as possible for fast and efficient measurement. However, it strongly depends on not only the collection order of the flow-stats but also the topological locations of lossy links in the multicast measurement route; the former one was investigated in the previous work but the latter has not been well explored. Therefore, in this paper, we develop a new dynamic scheme of building the multicast measurement route and controlling the collection order of flow-stats from switches, which leverages lossy link locations obtained in the recent past measurements in a repeated-measurement setting. The results of numerical simulation on real-world large-scale network topologies suggest the effectiveness and also the issues of the proposed lossy link location scheme.
- リンク情報
- ID情報
-
- DOI : 10.1109/ICOIN48656.2020.9016438
- ISSN : 1976-7684