Abstract: A system and a method for configuring resources over a network cloud are described. Attributes related to user roles i.e. categories of user roles, network cloud based services associated with each category, and a number of users associated with each category are received. Hardware capabilities and/or network capabilities corresponding to the attributes are determined from a mapping table stored in a repository. A service set capable of providing the hardware capabilities and the network capabilities is determined from the mapping table stored in the repository. The mapping table is based on previous implementations and instructions associated with one or more service sets. Resources are configured over the network cloud to implement the service set.

Abstract: Examples relate to seamless roaming across subnets. In an example, a system receives an indication that a client device has been authenticated at a first network device of a network. The system receives precursor keys and identification information of the client device, as generated from the authentication of the client device. The system determines second clusters, which are within a roaming domain of the cluster or the client device. The system receives one or more second precursor keys corresponding to the second clusters and distributes the second precursor keys to the corresponding second clusters. The system determines to create one or more tunnels among the second clusters and the cluster and provisions the tunnels to transmit data through the tunnels.

Abstract: Example implementation relates to a method of failure detection and seamless traffic switchover in a VPN system. A cluster of nodes exchange heartbeat messages to detect a failure at a first node in the cluster. When failure is detected at the first node, a master node transmits a failover message to a network end node connected to the first node. The failover message includes a list of active nodes to which traffic may be routed. The network end node updates its routing table based on the failover message and switches the traffic to a second node in the cluster of nodes.

Abstract: Data transfer for access points or switches in a cluster upon data tunnel failure is described. An example includes receiving uniform mapping information for a cluster including a bucket map mapping an active gateway and a standby gateway for each of multiple entries, the bucket map including mapping a first gateway node as a standby gateway and a second gateway node as an active gateway for an entry. Synchronized user information is received from the second gateway node including identification of a user indexed to the first entry. A message is received from a first AP or switch requesting activation of the user on the first gateway node as a standby gateway upon failure of a data tunnel between the first AP or switch and the second gateway node. The user is activated on the first gateway node.

Abstract: Systems and methods are provided for reducing WAN bandwidth consumption used by multicast for large scale software-defined branch deployments. In particular, a cloud-based multicast orchestrator may be implemented as part of an SD-WAN service. This cloud-based multicast orchestrator may orchestrate routes for multicast traffic between a multicast source and the various branches of the large scale software-defined branch deployment. This cloud-based multicast orchestrator may orchestrate routes for multicast traffic which reduce/optimize WAN bandwidth consumption. In combination with the cloud-based multicast orchestrator, examples may utilize a branch gateway hierarchy which designates one branch gateway a “leader” for a given multicast stream to further reduce WAN bandwidth consumption used by multicast.

Abstract: Examples relate to seamless roaming across subnets. In an example, a system receives an indication that a client device has been authenticated at a first network device of a network. The system receives precursor keys and identification information of the client device, as generated from the authentication of the client device. The system determines second clusters, which are within a roaming domain of the cluster or the client device. The system receives one or more second precursor keys corresponding to the second clusters and distributes the second precursor keys to the corresponding second clusters. The system determines to create one or more tunnels among the second clusters and the cluster and provisions the tunnels to transmit data through the tunnels.

Abstract: Some examples relate to distributing application classification entries to network devices. An example includes receiving, by a processing resource in a cloud computing system, an application classification entry for an application from respective network devices on a network. The application classification entry may comprise a given application identifier for identifying the application and control information for routing a network packet originating from the application. For the given application identifier, the processing resource may generate a consolidated set of application classification entries, based on the application classification entry received from respective network devices. The processing resource may then determine appropriate network devices to distribute the consolidated set of application classification entries.

Abstract: Example implementation relates to a method for establishing a dynamic VPN tunnel between branch gateway devices based on metric data. A branch orchestrator receives metric data from VPNC device. The metric data includes data center bandwidth and processor utilization of the VPNC device. The metric data is derived from the traffic being routed via the VPNC device. When the metric data associated with traffic between a first branch gateway device and a second branch gateway device is above a Service Level Agreement (SLA), a dynamic branch to branch VPN tunnel is established to route the traffic between the first branch gateway device and the second branch gateway device. The VPN tunnel between the branch gateways can be teared when the load at the VPNC device reduces.

Abstract: A system and a method of establishing seamless remote access VPN connections are described. For establishment of a VPN connection for a user device, a cluster leader of a cluster of controllers identifies an active controller and a standby controller, based on network load of each controller of the cluster of controllers. An active VPN connection is established between the user device and the active controller and a standby VPN connection is established between the user device and the standby controller. The standby VPN connection is utilized in place of the active VPN connection during failover of the active controller. Because information of an active session is regularly shared by the active controller to the standby controller, the standby controller can seamlessly resume the active session during failover of the active controller.

Abstract: Some examples relate to distributing application classification entries to network devices. An example includes receiving, by a processing resource in a cloud computing system, an application classification entry for an application from respective network devices on a network. The application classification entry may comprise a given application identifier for identifying the application and control information for routing a network packet originating from the application. For the given application identifier, the processing resource may generate a consolidated set of application classification entries, based on the application classification entry received from respective network devices. The processing resource may then determine appropriate network devices to distribute the consolidated set of application classification entries.

Abstract: Examples relate to maintaining consistent cluster data across a cluster in a network. A computing system may receive a first signature of a first state of the cluster data present at a leader gateway and a plurality of signatures of a plurality of states of the cluster data present at a plurality of member network devices of the cluster. The cluster may include a plurality of gateways including the leader gateway and a plurality of member gateways. The member network devices may include the plurality of member gateways and a plurality of interconnecting network devices. In response to determining that a signature of the plurality of signatures received from one of the member network devices is different from the first signature, the computing system may send a message to one of the plurality of gateways to update the cluster data at the member network device to represent the first state.

Abstract: A system and a method of establishing seamless remote access VPN connections are described. For establishment of a VPN connection for a user device, a cluster leader of a cluster of controllers identifies an active controller and a standby controller, based on network load of each controller of the cluster of controllers. An active VPN connection is established between the user device and the active controller and a standby VPN connection is established between the user device and the standby controller. The standby VPN connection is utilized in place of the active VPN connection during failover of the active controller. Because information of an active session is regularly shared by the active controller to the standby controller, the standby controller can seamlessly resume the active session during failover of the active controller.

Abstract: A system and a method for configuring resources over a network cloud are described. Attributes related to user roles i.e. categories of user roles, network cloud based services associated with each category, and a number of users associated with each category are received. Hardware capabilities and/or network capabilities corresponding to the attributes are determined from a mapping table stored in a repository. A service set capable of providing the hardware capabilities and the network capabilities is determined from the mapping table stored in the repository. The mapping table is based on previous implementations and instructions associated with one or more service sets. Resources are configured over the network cloud to implement the service set.

Abstract: Systems and methods are provided for receiving an indication that controllers and network devices belong to a site of a network, detecting the controllers and the network devices at the site, clustering the controllers to be deployed in the network, mapping the network devices to the controllers, and tunneling traffic from client devices accessing the network through the network devices to one of the controllers.

Abstract: Systems and methods are provided for performing a live upgrade process of one or more devices (e.g., access points (APs), etc.) in a large scale network environment. For example, the system can use neighbor path loss data between the APs. The neighbor path loss data may identify the signal strength between two points as an indication of the ability to transmit power and receive power from the individual APs. When the ability to transmit and receive power is present in one AP is unavailable the user device may connect to a second AP to avoid losing connectivity. This neighbor path loss data may be utilized to perform a live upgrade process of APs in parallel at multiple locations of the user site.

Abstract: Systems and methods are provided for providing a clustering solution and a Virtual Router Redundancy Protocol (VRRP) platform for a branch deployment of at least a first gateway and a second gateway. The VRRP platform can be associated with a VRRP Internet Protocol (IP) address and the first gateway configured as a VRRP master. A change in an operational status of the first gateway can be determined with the clustering solution. The second gateway can be caused to be configured, with the clustering solution, as a new VRRP master. Network traffic associated with the VRRP Internet Protocol (IP) address can be received. The second gateway can be caused to forward the network traffic.

Abstract: Systems and methods are provided for receiving an indication that controllers and network devices belong to a site of a network, detecting the controllers and the network devices at the site, clustering the controllers to be deployed in the network, mapping the network devices to the controllers, and tunneling traffic from client devices accessing the network through the network devices to one of the controllers.

Abstract: Example implementation relates to a method for establishing a dynamic VPN tunnel between branch gateway devices based on metric data. A branch orchestrator receives metric data from VPNC device. The metric data includes data center bandwidth and processor utilization of the VPNC device. The metric data is derived from the traffic being routed via the VPNC device. When the metric data associated with traffic between a first branch gateway device and a second branch gateway device is above a Service Level Agreement (SLA), a dynamic branch to branch VPN tunnel is established to route the traffic between the first branch gateway device and the second branch gateway device. The VPN tunnel between the branch gateways can be teared when the load at the VPNC device reduces.

Abstract: Systems and methods are provided for performing a live upgrade process of one or more devices (e.g., access points (APs), etc.) in a large scale network environment. For example, the system can use neighbor path loss data between the APs. The neighbor path loss data may identify the signal strength between two points as an indication of the ability to transmit power and receive power from the individual APs. When the ability to transmit and receive power is present in one AP is unavailable the user device may connect to a second AP to avoid losing connectivity. This neighbor path loss data may be utilized to perform a live upgrade process of APs in parallel at multiple locations of the user site.

Abstract: Example implementation relates to a method of failure detection and seamless traffic switchover in a VPN system. A cluster of nodes exchange heartbeat messages to detect a failure at a first node in the cluster. When failure is detected at the first node, a master node transmits a failover message to a network end node connected to the first node. The failover message includes a list of active nodes to which traffic may be routed. The network end node updates its routing table based on the failover message and switches the traffic to a second node in the cluster of nodes.