Overcoming Failures: Fault-tolerance and Logical Centralization in Clean-Slate Network Management

We investigate the design of a clean-slate control and management plane for data networks using the abstraction of 4D architecture, utilizing and extending 4D´s concept of logically centralized Decision plane that is responsible for managing network-wide resources. In this paper, a dynamically adaptable algorithm for assigning Data plane devices to a physically distributed Decision plane is presented, which enables a network to operate with minimal configuration and human intervention, while providing optimal convergence and robustness against failures. Our work is especially relevant in the context of ISPs and large geographically dispersed enterprise networks, where robust and scalable network-wide control of a large number of heterogeneous devices is desired.