Intent-driven composition of resource-management SDN applications

Abstract

As software-defined networking deployments mature, operators need to manage and compose multiple resource-management applications, such as traffic engineering and service chaining. Today such applications’ resource management algorithms run separately and composition approaches are output-driven, e.g., running each application on a statically provisioned slice of the network and then combining the flow rules output for each slice. Such approaches result in inefficient resource utilization and unfairness. Instead, we argue for intent-driven composition, where a unified resource optimization formulation is composed from applications’ high-level intents and the solution to this problem formulation is realized in the network. We design Chopin 1 , an intent-driven framework for composing SDN resource-management applications. Chopin’s design addresses key robustness challenges with regard to efficiency and fairness that arise in realizing such an intent-driven approach. We have integrated Chopin with the ONOS controller and show that it substantially improves efficiency and fairness over existing composition approaches.