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 virtual network function (VNF) controller (or module) instantiates two or more VNFs in a communication network to support a network service where the two or more VNFs include at least a first VNF and a second VNF. The VNF controller assigns a priority value to each VNF base on an overall network impact, a physical location of at least one network resource allocated to the respective VNF, a type of service to be implemented by the respective VNF and a customer impact based on how many customers would be using the respective VNF. The VNF controller monitors network resources allocated to each VNF. The VNF controller further determines the first VNF requires additional network resources and releases the network resources allocated to the second VNF based on respective priority values. The VNF controller further allocates the network resources released by the second VNF to the first VNF.

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: 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: In one embodiment, a method includes mapping micro-service network interfaces in a service network to service engine network interfaces for connecting micro-services to external endpoints, transmitting a request for communication with one of the external endpoints from one of the micro-services to a service broker operable to select one of the micro-service network interfaces for the communication with the external endpoint, and receiving a response from the service broker with the selected micro-service network interface. The service broker dynamically selects a service path for the communication based on a policy and independent from default network routes.

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: In one embodiment, a method includes mapping micro-service network interfaces in a service network to service engine network interfaces for connecting micro-services to external endpoints, transmitting a request for communication with one of the external endpoints from one of the micro-services to a service broker operable to select one of the micro-service network interfaces for the communication with the external endpoint, and receiving a response from the service broker with the selected micro-service network interface. The service broker dynamically selects a service path for the communication based on a policy and independent from default network routes.

Abstract: Systems, methods, and computer-readable storage media for network association to virtualize network devices using device passthrough. In some examples, a system can determine one or more respective configuration parameters associated with one or more network interfaces on the system. Based on the networking information, the system can map the one or more respective configuration parameters to each corresponding network interface from the one or more network interfaces, to yield mapped interfaces-parameters. Next, the system can record the mapped interfaces-parameters to a data object accessible by a virtual machine on the system. The system can then configure the virtual machine to process the mapped interfaces-parameters as boot time parameters while spawning and launch the virtual machine based on the mapped interfaces-parameters.

Abstract: A VNF controller (or module) instantiates two or more VNFs in a communication network to support a network service where the two or more VNFs include at least a first VNF and a second VNF. The VNF controller assigns a priority value to each VNF and monitors network resources allocated to each VNF. The VNF controller further determines the first VNF requires additional network resources and releases the network resources allocated to the second VNF based on respective priority values. The VNF controller further allocates the network resources released by the second VNF to the first VNF.

Abstract: In one embodiment, a method includes creating an active container and a standby container for a single supervisor of an operating system at a network device, instantiating instances for active Kernel Loadable Modules (KLMs) for servicing the active container, instantiating instances for standby KLMs for servicing the standby container, wherein one or more of the standby KLMs comprise upgraded versions of the active KLMs, and switching over from the active container to the standby container to perform an in-service upgrade of the KLMs for the operating system. An apparatus and logic are also disclosed herein.

Abstract: Method for assisting load-balancing of subscriber sessions in a distributed mobile gateway (e.g. PGW, SGW, ePDG, or TWAG) comprising a plurality of distributed gateway instances (DGIs) is disclosed. Method includes receiving an advertisement of GTP fully qualified tunnel endpoints identifications (F-TEIDs) allocated to a first plurality of subscriber sessions, the sessions assigned to a first DGI of the DGIs. The advertisement indicates the first DGI as a recipient for data provided to the gateway and related to any of the first plurality of sessions. Method further includes storing an association between the GTP F-TEIDs of the advertisement and the first DGI, receiving a packet comprising data related to a subscriber session of the first plurality of sessions, identifying a F-TEID for the data of the received packet, identifying from stored association, that the F-TEID is associated with the first DGI, and forwarding data of the packet to the first DGI.

Abstract: An example method for load-balanced IP pool distribution among a plurality of forwarding elements is provided. The method may include assigning a plurality of IP pool chunks to the plurality of forwarding elements, each IP pool chunk comprising a plurality of IP addresses, and receiving, from each of the forwarding elements, information indicative of a load on each forwarding element. The method may further include performing load balancing using the information indicative of the load on each forwarding element to determine whether re-assignment of the plurality of IP pool chunks is needed, and re-assigning at least one IP pool chunk of the plurality of IP pool chunks from a first forwarding element to a second forwarding element based on the load balancing.

Abstract: Systems, methods, and computer-readable storage media for network association to virtualize network devices using device passthrough. In some examples, a system can determine one or more respective configuration parameters associated with one or more network interfaces on the system. Based on the networking information, the system can map the one or more respective configuration parameters to each corresponding network interface from the one or more network interfaces, to yield mapped interfaces-parameters. Next, the system can record the mapped interfaces-parameters to a data object accessible by a virtual machine on the system. The system can then configure the virtual machine to process the mapped interfaces-parameters as boot time parameters while spawning and launch the virtual machine based on the mapped interfaces-parameters.

Abstract: Method for assisting load-balancing of subscriber sessions in a distributed mobile gateway (e.g. PGW, SGW, ePDG, or TWAG) comprising a plurality of distributed gateway instances (DGIs) is disclosed. Method includes receiving an advertisement of GTP fully qualified tunnel endpoints identifications (F-TEIDs) allocated to a first plurality of subscriber sessions, the sessions assigned to a first DGI of the DGIs. The advertisement indicates the first DGI as a recipient for data provided to the gateway and related to any of the first plurality of sessions. Method further includes storing an association between the GTP F-TEIDs of the advertisement and the first DGI, receiving a packet comprising data related to a subscriber session of the first plurality of sessions, identifying a F-TEID for the data of the received packet, identifying from stored association, that the F-TEID is associated with the first DGI, and forwarding data of the packet to the first DGI.

Abstract: An example method for load-balanced IP pool distribution among a plurality of forwarding elements is provided. The method may include assigning a plurality of IP pool chunks to the plurality of forwarding elements, each IP pool chunk comprising a plurality of IP addresses, and receiving, from each of the forwarding elements, information indicative of a load on each forwarding element. The method may further include performing load balancing using the information indicative of the load on each forwarding element to determine whether re-assignment of the plurality of IP pool chunks is needed, and re-assigning at least one IP pool chunk of the plurality of IP pool chunks from a first forwarding element to a second forwarding element based on the load balancing.

Abstract: In one embodiment, a method includes creating an active container and a standby container for a single supervisor of an operating system at a network device, instantiating instances for active Kernel Loadable Modules (KLMs) for servicing the active container, instantiating instances for standby KLMs for servicing the standby container, wherein one or more of the standby KLMs comprise upgraded versions of the active KLMs, and switching over from the active container to the standby container to perform an in-service upgrade of the KLMs for the operating system. An apparatus and logic are also disclosed herein.

Abstract: In one embodiment, a method includes generating at a network device comprising a virtual switch, a tenant record comprising tenant information for a context defined within the virtual switch, exporting the tenant record to a collector, monitoring network flow at the virtual switch, and exporting network flow data in a data record to the collector. The data record includes an identifier associating the data record with the context. An apparatus is also disclosed.

Abstract: A network appliance that is part of a distributed virtual switch collects network flow information for network flows passing through the network appliance. The network flow information is encapsulated into packets as a data record for transport. Network flow exporter type information is added to the network flow records configured to indicate that the packets are from a distributed exporter. An option template is sent to the network flow data collectors that includes a device identifier that is configured to uniquely identify the network appliance. The packets are exported to the network flow data collector. The network flow data collector uses the network flow exporter type information and the device identifier to associate the network flow information with the distributed virtual switch.

Abstract: In one embodiment, a method includes generating at a network device comprising a virtual switch, a tenant record comprising tenant information for a context defined within the virtual switch, exporting the tenant record to a collector, monitoring network flow at the virtual switch, and exporting network flow data in a data record to the collector. The data record includes an identifier associating the data record with the context. An apparatus is also disclosed.