Destination-aware Adaptive Traffic Flow Rule Aggregation in Software-Defined Networks

Abstract

In this paper, we propose a destination-aware adaptive traffic flow rule aggregation (DATA) mechanism for facilitating traffic flow monitoring in SDN-based networks. This method adapts the number of flow table entries in SDN switches according to the level of detail of traffic flow information that other mechanisms (e.g. for traffic engineering, traffic monitoring, intrusion detection) require. It also prevents performance degradation of the SDN switches by keeping the number of flow table entries well below a critical level. This level is not preset as a hard threshold but learned during operation by using a machinelearning based algorithm. The DATA method is implemented within a RESTful application (DATA App) which monitors and analyzes the ongoing network traffic and provides instructions to the SDN controller to adapt the traffic flow matching strategies accordingly. A thorough performance evaluation of DATA is conducted in an SDN emulation environment. The results show that-compared to the default behavior of common SDN controllers-the proposed DATA approach yields significant SDN switch performance improvements while still providing detailed traffic flow information on demand.

Q-DATA: Enhanced Traffic Flow Monitoring in Software-Defined Networks applying Q-learning

“…Based on the prediction result the Q-learning based Traffic Flow Matching Policy Creation module issues an optimum action on changing the traffic flow matching scheme. Note that this proposal partially inherits 1 our previous work [8] which is explained later on.…”

“…In our previous work, we proposed a destination-aware adaptive traffic flow rule aggregation solution named DATA [8] for adapting the number of flow entries in SDN switches according to the level of detail of traffic flow information that other mechanisms require and at the same time preventing SDN switch performance degradation.…”

“…The state of SDN switch i is defined as tuple s = ( f i , ∆ f i ) ∈ S i . In our previous study [8] we already discussed the reasons for choosing the tuple ( f i ,∆ f i ) as the representative for the state of an SDN switch.…”

“…Instead of always taking the best action according to the network state, the Q-learning agent will take some random actions, where the probability of a random decision is determined by the value of epsilon, . In its learning phase, the Q-learning agent first of all arbitrarily initializes the Q-table for all state-action pairs and afterwards updates it by using Equation 8. Accordingly, the agent acquires a trained or converged Q-table.…”

“…We observe that the switch starts getting overflowed or cannot handle new flow rules if the current total number of flows is around 3000 ( f cap i ) [6], [8]. Setting the idle_timeout value (after which the flow entries are removed) to 10 seconds, the safety threshold for the packet rate the switch can handle is 300 packets per second assuming that each packet belongs to a different traffic flow rule (worst case assumption).…”