Abstract: Techniques are presented herein for providing a persistent external Internet Protocol (IP) address for extra-cluster services. One example involves initiating, in a cluster, a first pod with a label that identifies a service. The first pod is configured to provide the service to one or more network entities outside the cluster. The first pod is assigned an IP address configured for communicating outside the cluster. A mapping of the service to the IP address is stored. In response to a determination that the service has been disrupted, a second pod is initiated in the cluster with the label that identifies the service. The second pod is configured to provide the service to the one or more network entities outside the cluster. Based on the mapping and the label that identifies the service, the IP address is assigned to the second pod.

Abstract: Two versions of a database can be held in two trees that have many of the same nodes. Both trees can be concurrently searched using recursive algorithms. A root node indicator indicates a root node for a tree search algorithm. The root node indicator can indicate a first root node of a first tree. A tree search algorithm can identify a record node in the first tree. Intermediate nodes between the record node and the first root node can be identified and retained nodes can be identified. A second root node and replacement intermediate nodes can be instantiated. A second tree that includes the second root node, the replacement intermediate node, and the retained nodes can be created. The root node indicator can be set to indicate the second root node after creating the second tree.

Abstract: Techniques and mechanisms for managing workloads in compute clusters comprising compute nodes by managing the workloads at the resource level of the compute clusters. For example, virtual service contexts (VSCs) may be defined where the VSCs represent service classes. Policies may be defined with respect to each service class. These service classes are dynamically constructed based on business needs. Hence there is natural requirement for a user to construct and rebalance the compute resources for these service classes dynamically. The policies may be related to resources of the compute clusters for executing workload units in the compute clusters. Resources of the compute clusters may be allocated to each service class. Each workload unit may be assigned to a one of the service classes based on the service context or type of workload unit. The workload units may then be executed by the compute clusters using the resources in accordance with the policies.

Abstract: Techniques for automatically claiming switches of a tenant computer network by a remote, cloud-based network controller. A first seed switch is manually claimed by a user by implementing the remote, cloud-based network controller. After claiming the seed switch a set of switches immediately connected with the seed switch are identified by Device Connector logic in the seed switch and immediately connected switches. Switches directly connected to those switches are then identified using Device Connector logic of the switches. This process is performed iteratively by identifying immediately connected switches until all of the switches are identified. All or a subset of the identified switches can then be claimed by the remote, cloud-based controller based on a response from the tenant network user.

Abstract: Techniques are presented herein for providing a persistent external Internet Protocol (IP) address for extra-cluster services. One example involves initiating, in a cluster, a first pod with a label that identifies a service. The first pod is configured to provide the service to one or more network entities outside the cluster. The first pod is assigned an IP address configured for communicating outside the cluster. A mapping of the service to the IP address is stored. In response to a determination that the service has been disrupted, a second pod is initiated in the cluster with the label that identifies the service. The second pod is configured to provide the service to the one or more network entities outside the cluster. Based on the mapping and the label that identifies the service, the IP address is assigned to the second pod.

Abstract: Techniques are presented herein for providing a persistent external Internet Protocol (IP) address for extra-cluster services. One example involves initiating, in a cluster, a first pod with a label that identifies a service. The first pod is configured to provide the service to one or more network entities outside the cluster. The first pod is assigned an IP address configured for communicating outside the cluster. A mapping of the service to the IP address is stored. In response to a determination that the service has been disrupted, a second pod is initiated in the cluster with the label that identifies the service. The second pod is configured to provide the service to the one or more network entities outside the cluster. Based on the mapping and the label that identifies the service, the IP address is assigned to the second pod.

Abstract: Two versions of a database can be held in two trees that have many of the same nodes. Both trees can be concurrently searched using recursive algorithms. A root node indicator indicates a root node for a tree search algorithm. The root node indicator can indicate a first root node of a first tree. A tree search algorithm can identify a record node in the first tree. Intermediate nodes between the record node and the first root node can be identified and retained nodes can be identified. A second root node and replacement intermediate nodes can be instantiated. A second tree that includes the second root node, the replacement intermediate node, and the retained nodes can be created. The root node indicator can be set to indicate the second root node after creating the second tree.

Abstract: Techniques are presented herein for providing a persistent external Internet Protocol (IP) address for extra-cluster services. One example involves initiating, in a cluster, a first pod with a label that identifies a service. The first pod is configured to provide the service to one or more network entities outside the cluster. The first pod is assigned an IP address configured for communicating outside the cluster. A mapping of the service to the IP address is stored. In response to a determination that the service has been disrupted, a second pod is initiated in the cluster with the label that identifies the service. The second pod is configured to provide the service to the one or more network entities outside the cluster. Based on the mapping and the label that identifies the service, the IP address is assigned to the second pod.

Abstract: Techniques and mechanisms for managing workloads in compute clusters comprising compute nodes by managing the workloads at the resource level of the compute clusters. For example, virtual service contexts (VSCs) may be defined where the VSCs represent service classes. Policies may be defined with respect to each service class. These service classes are dynamically constructed based on business needs. Hence there is natural requirement for a user to construct and rebalance the compute resources for these service classes dynamically. The policies may be related to resources of the compute clusters for executing workload units in the compute clusters. Resources of the compute clusters may be allocated to each service class. Each workload unit may be assigned to a one of the service classes based on the service context or type of workload unit. The workload units may then be executed by the compute clusters using the resources in accordance with the policies.

Abstract: A first provider edge (PE) device is configured to: receive a Label Distribution Protocol (LDP) MAC Flush message from a PE device via an input port; flush a routing table in response to the LDP MAC Flush message; determine whether the LDP MAC Flush message comprises a PE identifier corresponding to the PE device; generate a Topology Change Notification (TCN) message based on the LDP MAC Flush message when the LDP MAC Flush message comprises the PE identifier corresponding to the PE device; and output the TCN message.

Abstract: Techniques are described for supporting metro Ethernet “E-TREE” service over a packet-switched MPLS network, including a VPLS core, in a manner that allows a service provide to easily integrate with different types of technologies deployed by its various customers. Moreover, the techniques described herein provide increased flexibility with respect to the topology of the roots and leafs of the E-TREE service and, in particular, allow roots and leaf nodes to be coupled to a common router that provides access to the VPLS core. An NNI port of a PE router may process network traffic to provide E-TREE service to a bridged network having both leaf nodes and root nodes process and direct traffic between logical interfaces as changed next hops.

Abstract: In general, techniques are described for measuring packet data unit (PDU) loss in a L2 virtual private network (L2VPN) service, such as a VPLS instance. In one example of the techniques, provider edge (PE) routers that participate in the L2VPN measure known unicast and multicast PDU traffic at the service endpoints for the instance to determine unicast PDU loss within the service provider network. As the routers learn the outbound service (i.e., core-facing) interfaces and outbound local (i.e., customer-facing) interfaces for L2 addresses of customer devices that issue packets to the VPLS instance, the routers establish respective unicast transmit and receipt counters for the service endpoints that serve the customer devices. In another example, PE routers that participate in the L2VPN measure multicast PDU traffic at the service endpoints for the instance and account for internal replication by intermediate service nodes to determine multicast PDU loss within the service.

Abstract: The reliability of the connection of a client edge (CE) device to a core network may be improved using redundant provider edge (PE) devices. A first of the PE devices may monitor a connection to the core network, where the PE device acts as a root device in a set of devices that implement a spanning tree using a spanning tree protocol and where a second PE device in the set of devices additionally connects to the core network. The PE device may additionally detect failure of the connection of the PE device to the core network; and change, in response to the detected failure of the connection, a spanning tree protocol priority value of the device to a value having a lower priority than that of the second PE device.

Abstract: In general, techniques are described for enhanced learning in layer two (L2) networks. A first network device of the intermediate network comprising a control unit and an interface may implement these techniques. The control unit executes a loop-prevention protocol (LPP) that determines a bridge identifier associated with a second network device of the intermediate network, where the first and second network devices each couple to a first network. The LPP selects the second network device as a root bridge and detects a topology change that splits the first network into sub-networks. The interface then outputs a message to direct remaining network devices of the intermediate network to clear L2 address information learned when forwarding L2 communications. The message includes the bridge identifier determined by the loop-prevention protocol as the root bridge and directs these remaining network devices to clear only the L2 addresses learned from this bridge identifier.

Abstract: In general, techniques are described for measuring packet data unit (PDU) loss in a L2 virtual private network (L2VPN) service, such as a VPLS instance. In one example of the techniques, provider edge (PE) routers that participate in the L2VPN measure known unicast and multicast PDU traffic at the service endpoints for the instance to determine unicast PDU loss within the service provider network. As the routers learn the outbound service (i.e., core-facing) interfaces and outbound local (i.e., customer-facing) interfaces for L2 addresses of customer devices that issue packets to the VPLS instance, the routers establish respective unicast transmit and receipt counters for the service endpoints that serve the customer devices. In another example, PE routers that participate in the L2VPN measure multicast PDU traffic at the service endpoints for the instance and account for internal replication by intermediate service nodes to determine multicast PDU loss within the service.