This thesis investigates the design of hierarchical software-defined networks with improved scalability and service integration. Software Defined Networking (SDN) aims to bring flexibility and intelligence into networks by moving the network stack into logically centralized, programmable control planes. While distribution schemes and virtualization have addressed control plane scalability and network stack coexistence, they do so separately as unrelated subjects, and no unified solution exists. This work addresses the intersection of the two aspects through a little-explored control plane distribution method in which controllers are organized into a hierarchy based on service characteristics. Specifically, this thesis addresses 1) the evolution of SDN control plane architectures, focusing on distributed SDNs based on the OpenFlow SDN control protocol, and 2) the design, implementation, and evaluation of an architecture for a hierarchical OpenFlow control plane. A layered model for control plane architecture is presented as a reference for the design. The widely-adopted Floodlight SDN framework is used to implement a prototype. The prototype is implemented as a series of modules that extend the base platform to use an OpenFlow-based inter-controller protocol. This protocol enables controllers to configure one another based on their positions in the hierarchy, enabling them to coordinate event handling by the network services at each controller. This prototype is evaluated on the wired sandboxes of the ORBIT network testbed, with focus on control plane overhead and scalability. A custom OpenFlow client is used to measure processing overheads for several control plane topologies. Scalability is evaluated with respect to event processing rate at the lowest tier, and the amount of requests received at the upper tiers, of a sample hierarchy. Each controller involved in event handling is found to add at most about 0.46ms to overhead per packet, discounting network link delays. A comparison with the stock controller reveals that the control plane scales in event processing capacity with the number of data plane-facing controllers, and that the volume of requests at higer tiers are coupled to the types of events requested. We finish with remarks on potential future improvements to the control plane design.
Subject (authority = RUETD)
Topic
Electrical and Computer Engineering
Subject (authority = ETD-LCSH)
Topic
Software engineering
Subject (authority = ETD-LCSH)
Topic
Computer software--Development
RelatedItem (type = host)
TitleInfo
Title
Rutgers University Electronic Theses and Dissertations
Rutgers University. Graduate School - New Brunswick
AssociatedObject
Type
License
Name
Author Agreement License
Detail
I hereby grant to the Rutgers University Libraries and to my school the non-exclusive right to archive, reproduce and distribute my thesis or dissertation, in whole or in part, and/or my abstract, in whole or in part, in and from an electronic format, subject to the release date subsequently stipulated in this submittal form and approved by my school. I represent and stipulate that the thesis or dissertation and its abstract are my original work, that they do not infringe or violate any rights of others, and that I make these grants as the sole owner of the rights to my thesis or dissertation and its abstract. I represent that I have obtained written permissions, when necessary, from the owner(s) of each third party copyrighted matter to be included in my thesis or dissertation and will supply copies of such upon request by my school. I acknowledge that RU ETD and my school will not distribute my thesis or dissertation or its abstract if, in their reasonable judgment, they believe all such rights have not been secured. I acknowledge that I retain ownership rights to the copyright of my work. I also retain the right to use all or part of this thesis or dissertation in future works, such as articles or books.