Abstract: A method for balancing virtual machine loads is provided. The method comprises: monitoring load information of each of physical hosts and virtual machines operating on said physical hosts; calculating load index of each of said physical hosts and said virtual machines operating on said physical hosts respectively according to said load information; based on load index of each physical hosts, determining a source physical host on which any virtual machine needs to be migrated so as to reduce the load index of the source physical host; determining a target virtual machine to be migrated on the source physical host based on load index of each virtual machines on the source physical host and respective affinity factor indicating a degree of dependency of a virtual machine on a physical host on which it resides; and migrating the target virtual machine into a destination physical host different than the source physical host.

Abstract: Based on an example, a core backbone (CB) device receives via a second-level stacking port connected with a source port extender (PE) device, a packet containing a source port number and a local identifier of a source forwarding chip. The CB device transmits the packet via a second-level stacking port connected with a destination PE device, wherein the packet contains the source port number, a global identifier of the source forwarding chip, a local identifier of a destination forwarding chip and a destination port number.

Abstract: The present disclosure provides a method for backing up a link state advertisement (LSA) for use in a distributed device comprising a main control board and an interface board. The main control board comprises a primary board and a standby board. The method comprises: receiving by the standby board LSA related information sent by the primary board; and generating by the standby board an LSA according to the LSA related information.

Abstract: According to an example, a virtual next-hop is created and a destination identity corresponding to each virtual next-hop is stored. Each virtual next-hop comprises one best next-hop and one second-best next-hop, and a next-hop of the best path of the destination identity corresponding to a virtual next-hop is the best next-hop of the virtual next-hop, and a next-hop of the second-best path of the destination identity corresponding to the virtual next-hop is the second-best next-hop of the virtual next-hop. Path forwarding may be performed using the best next-hop of each virtual next-hop. If the best next-hop of a virtual next-hop is determined to be failed, path forwarding may be performed using the second-best next-hop of the virtual next-hop.

Abstract: A method for constructing a network enabling layer-2 interconnection of data centers includes: a current data center registers at a neighbor server when the current data center gets online, obtains information of all registered data centers, generates a virtual bridge port for each registered data center; after the current data center comes online, the current data center obtains information of all newly registered data centers according to a pre-set rule, generates a virtual bridge port for each newly registered data center; establishes a tunnel to each registered data center by using the virtual bridge port corresponding to the registered data center, and transmit data via the tunnel.

Abstract: A method and an edge device for learning a Medium Access Control (MAC) address are provided. According to an example, based on a request message from a first node to a second node, a session cache item is created in the edge device. The session cache item includes: IP addresses of the first and second nodes, and a first relationship between a MAC address of the first node and an output port corresponding to the first node. Based on a response message from the second node to the first node, the session cache item is updated to include a second relationship between a MAC address of the second node and an output port corresponding to the second node. The first and second relationships are then provided to a data plane of the edge device to facilitate data packet forwarding between the first node and the second node.

Abstract: According to an example of the present disclosure, a service template providing a service may be found according to type of the service. A resource zone matching the service template may be found from a resource pool. A network resource matching a service unit in the service template may be found from the resource zone. A network parameter configured for the service unit may be sent to the network resource. Usage state information of the network resource may be updated.

Abstract: A first migration instruction is received from a management device, new authentication information is configured on a routing device according to the first migration instruction, and an authentication direction of the new authentication information is configured as a receiving direction. A second migration instruction that is sent by the management device after determining that all adjacent routing devices have configured the new authentication information is received, an authentication direction of original active authentication information is configured as a receiving direction and the authentication direction of the new authentication information is configured as the receiving direction and a sending direction.

Abstract: In a VPLS network, a first PE in a VPN instance sends topology information and a bridging MAC address of the first PE to another PE in the VPN instance, receives and stores topology information and bridging MAC addresses from the another PE. The first PE further generates a CE MAC table in the VPN instance according to the stored topology information and bridging MAC addresses of the another PE; calculates a unicast minimal spanning tree by taking the first PE as a root, generates a unicast MAC table in the VPN instance according to the unicast minimal spanning tree. Each PE further selects one of the PEs in the VPN instance as a root bridge, calculates a broadcast minimal spanning tree by taking the root bridge as a root, and generates a broadcast MAC table in the VPN instance according to the broadcast minimal spanning tree.

Abstract: The disclosure relates to a method of routing for an Autonomous System AS in a communication system, in which the AS comprises a plurality of network devices including a first network device and a second network device. The first network device distributes routing information to a forwarding table by calculating a first route from the first network device to the second network device. The first network device receives information of a destination network device of the communication system from the second network device, and forms a second route from the second network device to the destination network device to be stored in the forwarding table using the received information of the destination network device. When the first route fails, the first network device calculates a third route from the first network device to the second network device and updates the routing information in the forwarding table according to the third route.

Abstract: An access point (AP) for an indoor antenna distributed deployment environment provides an access service for a wireless terminal. The AP includes N radio frequency (RF) processing units and N*M RF switches (RFSs). Each RFS includes a public port and at least two branch ports. The at least two branch ports respectively connect with at least two antennas deployed in different indoor areas via RF feeders. The RF processing unit includes M RF chains. Each RF chain is connected to a public port of an RFS. The RF processing unit is to transmit a packet to the wireless terminal, based on a predetermined antenna combination for the wireless terminal.

Abstract: According to an example, in a method and apparatus for notifying a LSP, the downstream LSR detecting the link failure on the work LSP sends a SF message on the direction of the work LSP, so that the upstream LSR detecting the link failure on the work LSP receives the SF message and sends the packet via the protection LSP according to the SF message.

Abstract: A method of zero configuration for a virtual distributed device in a distributed network is disclosed. A plurality of peer devices are added as virtual devices to a network device. Configuration information is imported to the network device. A peer group comprising the plurality of peer devices, and corresponding topology of the peer group, are established in the network device. A connection between the network device and one of the plurality of peer devices is established via an uplink switch, the plurality of peer devices being connected to one another by virtual links. The network device distributes the configuration information to the plurality of peer devices in accordance with the established topology via the peer device connected to the uplink switch.

Abstract: This disclosure provides methods and devices for Graceful Restart (GR) in a network with multiple GR Restarters that include a first GR Restarter and at least one second GR Restarter corresponding to the first GR Restarter. After receiving routes from all normal neighbor network devices of the first GR Restarter, the first GR Restarter performs route selection to determine first selected routes, and sends the first selected routes to the at least one second GR Restarter. After receiving routes from the at least one second GR Restarter, the first GR Restarter performs route selection to determine second selected routes, updating a local Forwarding Information Base (FIB) at the first GR Restarter with the second selected routes, and sends the second selected routes to all neighbor network devices of the first GR Restarter. The first GR Restarter directly connects with the at least one second GR Restarter corresponding to the first GR Restarter.

Abstract: Providing a VLAN virtual interface of a default VLAN as a VLAN virtual interface of a layer-3 Ethernet interface provided on an apparatus, assigning an interface MAC address for the layer-3 Ethernet interface, setting a layer-3 interface entry including at least the interface MAC address of the layer-3 Ethernet interface and a VLAN ID of the default VLAN, providing a VLAN virtual interface of a layer-3 Ethernet sub-interface provided on the apparatus based on a binding VLAN of the layer-3 Ethernet sub-interface, assigning an interface MAC address to the layer-3 Ethernet sub-interface, setting a layer-3 sub-interface entry including at least the interface MAC address of the layer-3 Ethernet sub-interface and a VLAN ID of the binding VLAN, receiving a packet from one interface of the apparatus, and performing layer-3 forwarding for the packet if the layer-3 interface entry or the layer-3 interface entry matching with a destination MAC address and a VLAN ID of the packet.

Abstract: According to an example, in a method for MAC address distribution by an apparatus, a MAC address is received in the apparatus, in which the apparatus includes a data store on which is stored a local slice set containing a plurality of slices, and in which each of the plurality of slices is assigned a respective slice identifier. An address value for the received MAC address may be determined and a correlation between the determined address value and one of the slice IDs of the plurality of slices may be identified. In addition, the received MAC address may be stored in the slice having the slice identifier identified as being correlated to the determined address value.

Abstract: A method and an edge device for learning a Medium Access Control (MAC) address are provided. According to an example, based on a request message from a first node to a second node, a session cache item is created in the edge device. The session cache item includes: IP addresses of the first and second nodes, and a first relationship between a MAC address of the first node and an output port corresponding to the first node. Based on a response message from the second node to the first node, the session cache item is updated to include a second relationship between a MAC address of the second node and an output port corresponding to the second node. The first and second relationships are then provided to a data plane of the edge device to facilitate data packet forwarding between the first node and the second node.

Abstract: According to an example, a virtual port is configured for a current virtual device (VD) to interconnect the current VD with other VDs. The virtual port may be a layer 3 port and route learning and advertising of port information may be performed through the virtual port.

Abstract: When updated routing information is advertised, attributes of the updated routing information are advertised in a first UPDATE message and prefixes of the updated routing information are advertised in a second UPDATE message.

Abstract: According to an example, a method for forwarding an Ethernet packet may include receiving an Ethernet packet having a destination MAC address that is a MAC address of a gateway, selecting a port from multiple ports, wherein each of the ports of the multiple ports connects to one member device of a VRRP router that is working as the gateway; and sending the Ethernet packet via the selected port.