The (surprising) computational power of the SDN data plane

Abstract

A software defined network (SDN) separates the centralized control plane from the distributed data plane. This approach simplifies control logic at the cost of a heavy burden on the software-based controller and potential long reaction time to data plane events. One solution to this problem is to distribute control logic to multiple controllers spread across the network. Such a solution, however, requires additional mechanisms to enforce correctness properties (e.g., consistency) among the controllers and it still does not fully eliminate latency, as controller decisions happen in software. In this paper, we explore a novel approach to this problem: configuring the rules used by the data plane switches to allow these switches to effectively handle latency-sensitive network management tasks without the direct intervention of the control plane. We are not suggesting to add distributed control logic capability to the switches, we are instead exploring the feasibility of encoding such logic using the standard forwarding rules already available to these devices. To this end, we formally model a network of SDN switches, and then prove using tools from computability theory that such systems are capable of simulating polynomial space Turing Machines, indicating a surprising amount of computational power.

On scaling software-Defined Networking in wide-area networks

“…Further, Ref. [32] theoretically proves that under such a standardized control scheme, the potential computation capacity in the dataplane is enough for flexible control.…”

Polynomial-Time What-If Analysis for Prefix-Manipulating MPLS Networks

“…The polynomial time algorithm from [29] relies on the restriction of a finite set of header symbols, and contrary to our work, this does not allow to model networks with unboundedly many headers, like MPLS and segment routing networks. The complexity question is also related to the recent studies on the (sometimes surprising) computational power (and hence complexity) of different networks: e.g., Peresini and Kostic [30] showed how to simulate a Rule 110 cellular automaton in a general network model, and Newport and Zhou [31] highlighted the computational power of SDNs.…”