OnScalabilityofSoftware-DefinedNetworking.pptVIP

On Scalability ofSoftware-Defined Networking Author: Soheil Hassas Yeganeh, Amin Tootoonchian, and Yashar Ganjali, University of Toronto Publisher: IEEE Communications Magazine 2013 Presenter: Sih-An Pan, Date: 2013/10/2 Introduction Moving the control function out of data plane elements is the common denominator among software-defined networking (SDN) proposals in the research community. The common perception that control in SDN is centralized leads to concerns about SDN scalability and resiliency. Advantages of SDN High Flexibility Vendor-agnostic Programmability Centralized Network View Software-Defined Network Centralized Controller A. Tootoonchian et al., “On Controller Performance in Software-Defined Networks,” Proc. USENIX HotICE’12 Simple modifications to the NOX controller Boosts its performance by an order of magnitude on a single core DevoFlow (ACM Sigcomm 2011) With support from ASIC Short-lived flows are handled in the data path Only larger flows are forwarded to the controller Reducing the load on controller Distributed Controller Design Tradeoff Strictly Consistent Centralized Viewwill hinder response time and throughputas the network scales It is NOTalways feasible to achieve strong consistency while maintaining availability and partition tolerance Selecting an apt consistency level is an important design trade-off in SDN Distributed Controller HyperFlow(INM Conf. 2010) Synchronizes network state among multiple controller Giving the controller an illusion of control over the whole network Flow Initiation Overhead Resiliency to Failures Opportunities and Challenges Testing and Verification Without the right tools in place, troubleshooting becomes a major problem for networks as they grow. Extensibility Current SDN implementations take a pragmatic approach in supporting only well-known popular protocols that are readily supported by most equipment and chip vendors API and function for new protocol OpenFlow 將網路分為資料層 (Data Plane) 與控制層(Contro