高存活OpenFlow網路的路徑規劃與植基於軟體的快速故障偵測

在本文中，我們主要研究OpenFlow網路的高存活率和可擴充性。在測試過一些商業用的OpenFlow交換機後，我們發現這些交換機在故障偵測花費了很長的時間，導致無法在OpenFlow網路中實現快速恢復。為了解決這個問題，我們在文中的第一部分，提出一個軟體的網路故障恢復方法，透過提供可經過網路上所有鏈結的多條monitoring circles，除了可以在短時間內偵測到故障，並且明確指出故障鏈結的位置，而且這個方法可被應用在out-of-band或in-band的OpenFlow網路上，另外，我們引用min-max k-Chinese postman problem提出一個演算法，透過這個演算法，我們可以得到整個網路上所需設置的monitoring circles。文中的第二部分，主要在於研究路徑規劃的問題，在OpenFlow網路中，controller必須負責控制和管理整個網路，而透過中央控管的機制雖然可以有效的使用網路資源，但是卻有可能因為網路太大使得controller因為無法即時處理所有的需求導致網路效率降低，所以我們提出一個預先設置路徑的方法來減輕controller的負擔。這個方法的基本做法就是將路徑上除了入口和出口以外的交換機的flow entry都先設置好，當有flow進到網路中，controller只需設置該路徑的入口和出口交換機的flow entry，便可完成路徑的設置，因為整個網路的路由表已事先計算好加上大部分的交換機已完成設置，所以從允許flow進入到完成路徑的設置就變得非常簡單。另外，在路徑規劃方面還加入負載平衡、網路存活率、快速恢復和flow entry的總消耗量當作考量，最後，參考ISF (interface specific forwarding)提出一個路徑規劃的方法來求出我們所需的路由表，而這個路徑規劃的問題可被寫成一個integer linear programming problem，再經過heuristic演算法來得到一個次佳解。最後，藉由廣泛的實驗和模擬來計算我們提出的方法之效能，而實驗結果指出我們提出的軟體偵測故障的方法對controller的負擔極小，而且在高負載的網路中仍能做到快速恢復，另外，在模擬結果可看到我們引用ISF提出的預先設置路徑的方法可以減少controller的負擔並且只花費少量的總頻寬消耗量下就可以做到100%單一鏈結故障存活率。

關鍵字

快速故障偵測；快速恢復；允入控制；路徑規劃；可擴充OpenFlow網路

並列摘要

In this paper, we investigate survivable and scalable OpenFlow network design problem. We have examined several commercial OpenFlow switches and find that the long failure detection time makes an OpenFlow network unable to meet the requirement for fast failure recovery. To resolve this problem, we propose a software based failure recovery scheme. Through provisioning monitoring cycles to cover all links in the network, the proposed scheme can detect a failure and pinpoint the failure location in short time. It can be applied for both out-of-band and in-band controlled OpenFlow networks. An algorithm based on the solution of min-max k-Chinese postman problem is proposed to realize the planning of monitoring cycles. In the second part of this work, we investigate the pre-planned path layout problem. In an OpenFlow network, the controller is responsible to control and manage the whole network. Although such central control paradigm is easier to achieve efficient usage on network resources, the controller tends to be the bottleneck when the network size is large. We use pre-provisioned paths to mitigate the burden on controller. The basic idea is to setup some background paths in the network. When a flow coming into the network, the controller only needs to setups flow entries at the ingress and egress switches. Since routing paths are pre-planned, the whole admission and connection setup processes are simplified. Jointly considering load balancing, network survivability, fast failure recovery, and flow entry consumption are the major challenges in this problem. We propose a path preplanning scheme that takes advantage of interface specific forwarding (ISF) to resolve this problem. The path preplanning problem is formulated as an integer linear programming problem. We present an optimization based heuristic algorithm to obtain a near optimal solution. Extensive experiments and simulations are performed to evaluate the performance of the proposed schemes. The experimental results indicate that the proposed software based failure detection scheme generates very limited overhead on the OpenFlow controller. It can be used to achieve fast failure recovery even when the network is in heavy load state. The simulation results also show that the proposed ISF based path preplanning scheme can reduce controller load and is able to achieve 100% survivability against any single link failure with low total bandwidth consumption.

並列關鍵字

Fast Failure Detection ； Fast Failure Recovery ； Admission Control ； Path Provisioning ； Scalable OpenFlow Network

參考文獻

[1] OpenFlow Switch Specification 1.3.2:

[4] Sachin Sharma, Dimitri Staessens, Didier Colle, Mario Pickavet and Piet Demeester, “Enabling Fast Failure Recovery in OpenFlow Networks,” in Proc. DRCN, 2011.

[5] Dimitri Staessens, Sachin Sharma, Didier Colle, Mario Pickavet and Piet Demeester, “Software Defined Networking: Meeting Carrier Grade Requirement,” in Proc. IEEE LANMAN, 2011.

[6] James Kempf, Elisa Bellagamba, András Kern, Pontus Sköldström , “Scalable Fault Management for OpenFlow,” in Proc. IEEE ICC, 2012.

[7] Sachin Sharma, Dimitri Staessens, Didier Colle, Mario Pickavet and Piet Demeester, “Fast failure recovery for in-band OpenFlow netwoks,” in Proc. DRCN, 2013

國際替代計量

高存活OpenFlow網路的路徑規劃與植基於軟體的快速故障偵測

未授權

主題瀏覽