在軟體定義網路下節省頻寬

Improve Bandwidth Consumption in Distributed SDN Control Platform

鄧鈞謙(Chun-Chien Teng)

指導教授 ： 周承復

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2018年)

https://doi.org/10.6342/NTU201800679

摘要

軟體定義網路(SDN)是近年來越來越興起的技術，它將控制介面跟傳輸介面分開來，如此一來讓我們可以更加方便的對網路做管理，如果我們固定交換器跟控制器的連線方式，我們可以發現到說可能會造成有一些控制器過載，但是有一些控制器卻沒有被完全利用，為了要使控制器間的負擔平均分配，以及能快速地回應交換器的訊息，很多研究專注於如何適當動態的的分配控制器，較少研究專注在如何分配備份控制器，然而在像是資料中心之類的地方，頻寬是不能被浪費且珍貴的資源，因此為了要最小化頻寬的消耗，我們提出一個最佳化的問題，而且我們發現到要找到這問題最佳解是很困難的，因此我們提出一個使用K-means跟coalitional game的高效率的演算法去解決這問題，經實驗證明，我們提出的方法跟最佳解找到的答案非常相近，而且使用我們提出的演算法還可以避免全部控制器都掛。

關鍵字

軟體定義網路 ； 容錯

並列摘要

In recent year, Software defined networking is becoming increasingly prevalent because of its programmability that enables centralized network configuration and management. It is obvious that static assignment will cause some controller overloaded while some controller is not fully utilized and increase response time. To achieve load-balancing among controllers and low response time, a lot of works focus on how to dynamicly assign master controller and less works focus on slave controller assignment. However, bandwidth is scarce in some place such as data center so we should take bandwidth consumption into consideration when we are assigning slave controller. We formulate bandwidth consumption as mix integer linear programming model and figure out that it is very diffifcult to solve it. As a result, we propose a algorithm to use k-means and coalitional games to solve it efficiently. The simulation result shows that result of our proposed algorithm can not only be very closed to optimal result but also isolate failure effect.

並列關鍵字

Software defined network ； Failover

參考文獻

[2] M. Al-Fares, A. Loukissas, and A. Vahdat. A scalable, commodity data center network architecture. In ACM SIGCOMM Computer Communication Review, volume 38, pages 63–74. ACM, 2008.

[3] Y. Chen, S. Alspaugh, and R. Katz. Interactive analytical processing in big data systems: A cross-industry study of mapreduce workloads. Proceedings of the VLDB Endowment, 5(12):1802–1813, 2012.

[5] R. Cohen, L. Katzir, and D. Raz. An efficient approximation for the generalized assignment problem. Information Processing Letters, 100(4):162–166, 2006.

[11] F. Liu, J. Guo, X. Huang, and J. C. Lui. eba: Efficient bandwidth guarantee under traffic variability in datacenters. IEEE/ACM Transactions on Networking, 25(1):506–519, 2017.

[13] L. F. M¨uller, R. R. Oliveira, M. C. Luizelli, L. P. Gaspary, and M. P. Barcellos. Survivor: an enhanced controller placement strategy for improving sdn survivability. In Global Communications Conference (GLOBECOM), 2014 IEEE, pages 1909–1915. IEEE, 2014.