Abstract: A method is provided that is performed by a first node of a fat tree network for management of a plurality of traffic flows having differentiated service requirements. The method includes receiving, from a simulation model or a testbed environment, joint observations for the traffic flows. The traffic flows correspond to routing paths comprising different combinations of at least one leaf node, at least one spine node, and at least one super spine node of the fat tree network per routing path. The method further includes identifying, with a first reinforcement learning model, a first action to take to reduce or prevent congestion at the first node. The first action includes a reconfiguration of the first node for an identified routing path. The method further includes outputting, to a controller node, the reconfiguration of the first node.

Abstract: A computer implemented method is disclosed for generating a policy for managing a configuration of internal components of a communication network node. The method includes obtaining performance data and generating a model of the communication network node The method further includes using the model to generate a first data set; and extracting, from the first data set: a set of conditional probabilities of operational state transition for the communication network node; and a set of conditional probabilities of changes in observed measure of performance for the communication network node. The method further comprises includes combining the extracted sets of conditional probabilities with a reward function to form a configuration model and generating a solution to the configuration model including a policy that is operable to propose a change in configuration of an internal component of the communication network node based on an observed measure of performance of the communication network node.