In one example, a location of a potential bottleneck of network traffic in a network is identified. Based on the location of the potential bottleneck, a first QoE inference model is selected from a plurality of respective QoE inference models. The respective QoE inference models are each trained to infer a respective QoE of the network traffic based on one or more respective network traffic metrics generated by monitoring the network traffic at a respective location in the network. One or more first network traffic metrics of the one or more respective network traffic metrics are generated by monitoring the network traffic at a first respective location. The one or more first network traffic metrics are provided to the first QoE inference model to infer a first respective QoE.

Various embodiments providing for an indicator (termed the “Traffic Category Indicator,” TCI) to be encoded into packets, different values of which can be used, e.g., to distinguish Traffic Engineered (TE) packets and non-TE packets. In an example embodiment, the TCI can be used, e.g., to configure a network node to implement different packet queues, on each link, for TE packets and non-TE packets. In embodiments corresponding to the DiffServ TE paradigm, a node can be configured to implement different queues within each Forwarding Class for each link, said different queues distinguished by different respective TCI values. Example benefits of TCI include, but are not limited to fate separation of TE and non-TE packets in a node. The TCI concept can beneficially be applied to different packet-switching technologies supporting Source Routing, such as the IP, MPLS, Ethernet, etc.

Disclosed is a system comprised of different points-of-presence (“PoPs”) that provide software defined hybrid private and public networking for different entity sites. A system PoP may establish and/or maintain different network circuits for accessing different destinations or services. The system PoP may receive a tunnel creation request from a particular site, may obtain entity-defined policies that are specified for the particular site, and may establish a particular network tunnel to the particular site. The system PoP may then route egress traffic with a first classification through a first network circuit with a different quality of service (“QoS”) than egress traffic with a second classification according to the policies. Similarly, the system PoP may route ingress traffic arriving over the first network circuit through the particular network tunnel with a different QoS than ingress traffic arriving over the second network circuit according to the policies.

Disclosed is a system comprised of different points-of-presence (“PoPs”) that provide software defined hybrid private and public networking for different entity sites. A system PoP may establish and/or maintain different network circuits for accessing different destinations or services. The system PoP may receive a tunnel creation request from a particular site, may obtain entity-defined policies that are specified for the particular site, and may establish a particular network tunnel to the particular site. The system PoP may then route egress traffic with a first classification through a first network circuit with a different quality of service (“QoS”) than egress traffic with a second classification according to the policies. Similarly, the system PoP may route ingress traffic arriving over the first network circuit through the particular network tunnel with a different QoS than ingress traffic arriving over the second network circuit according to the policies.

Improved network traffic flow processing techniques are described. In a network device providing multiple processing planes, each processing plane comprising multiple processing units, techniques are described that take advantage of flow affinity/locality principles such that the same processing component of a processing plane, which previously performed processing for a network flow, is used for performing subsequent processing for the same network flow. This enables faster processing of network traffic flows by the network device. In certain implementations, the techniques described herein can be implemented in a network virtualization device (NVD) that is configured to perform network virtualization functions.

For efficiently handling network traffic via a fixed access, classification of UL data traffic in a communication device may be accomplished in a reflective mode by detecting identifiers of outgoing uplink data packets which are complementary to identifiers of incoming downlink data packets. The downlink data packets are already assigned to the traffic classes. The outgoing uplink data packets carrying the complementary identifier are assigned to the same traffic class as the incoming downlink data packets. For this purpose, the communication device is provided with a traffic classificator. In the reflective mode, the traffic classificator locally generates uplink packet classification rules by monitoring received downlink data packets.

For efficiently handling network traffic via a fixed access, classification of UL data traffic in a communication device may be accomplished in a reflective mode by detecting identifiers of outgoing uplink data packets which are complementary to identifiers of incoming downlink data packets. The downlink data packets are already assigned to the traffic classes. The outgoing uplink data packets carrying the complementary identifier are assigned to the same traffic class as the incoming downlink data packets. For this purpose, the communication device is provided with a traffic classificator. In the reflective mode, the traffic classificator locally generates uplink packet classification rules by monitoring received downlink data packets.

Some embodiments provide a method for forwarding data messages between edge nodes that perform stateful processing on flows between a logical network and an external network. At a particular edge node, the method receives a data message belonging to a flow. The edge nodes use a deterministic algorithm to select one of the edge nodes to perform processing for each flow. The method identifies a first edge node to perform processing for the flow in a previous configuration and a second edge node to perform processing for the flow in a new configuration according to the algorithm. When the first and second edge nodes are different, the method uses a probabilistic filter and a stateful connection tracker to determine whether the flow existed prior to a particular time. When the flow did not exist prior to that time, the method selects the second edge node for the received data message.

Some embodiments provide a method for forwarding data messages between edge nodes that perform stateful processing on flows between a logical network and an external network. At a particular edge node, the method receives a data message belonging to a flow. The edge nodes use a deterministic algorithm to select one of the edge nodes to perform processing for each flow. The method identifies a first edge node to perform processing for the flow in a previous configuration and a second edge node to perform processing for the flow in a new configuration according to the algorithm. When the first and second edge nodes are different, the method uses a probabilistic filter and a stateful connection tracker to determine whether the flow existed prior to a particular time. When the flow did not exist prior to that time, the method selects the second edge node for the received data message.

A communication equipment receives a plurality of network sessions and includes a memory and a processor. The memory stores a plurality of software modules. The processor is connected to the memory and configured to implement the following steps. A network session inferring step is performed to execute an inference software module, and the inference software module processes at least one network packet of each of the network sessions and a packet characteristic module according to a machine learning algorithm to infer a priority level. An adaptive priority list establishing step is performed to execute a classification software module, and the classification software module establishes an adaptive priority list according to the priority levels of the network sessions. The communication equipment transmits the network packets of each of the network sessions to a network according to the adaptive priority list so as to set QoS of the network sessions.