A software-defined wide area network (SD-WAN) controller in a cloud service delivery architecture may encounter a number of challenges including, for example, timeouts, download or push failures, etc. To address challenges of these types techniques are presented herein that support a holistic solution. Aspects of the solution comprise, among other things, understanding a device’s ability to process configuration in a real-time network and staggering a configuration push based on, for example, a device’s ability to process and consume configurations. Aspects of the solution employ, for example, the use of bandwidth knowledge to force a centralized network management system control connection to come up on a specific wide area network (WAN) interface that has higher bandwidth availability, the collection of network characteristics to profile a device’s behavior for receiving different types of policies from a SD-WAN controller, the scheduling of SD-WAN customer premise equipment (CPE) devices for configuration delivery in an ordered fashion such that the devices that are chosen are the ones with the least WAN congestion, the profiling of device connectivity patterns to provide guaranteed service-level agreement (SLA) commitments for policy enforcement, the possible use of "dummy templates" for diagnostic purposes, etc.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Sundararajan, Balaji; Lynch, Brandon; Chandrasekaran, Gayathri; Gatus, Reuel; and Aziz, Zaheer, "ADAPTIVE MODELS FOR NEXT GENERATION WIDE AREA NETWORKS", Technical Disclosure Commons, (December 22, 2020)