| 摘要 |
A mesh network, operating as a virtual Ethernet switch, includes multiple nodes operating as Mesh Network Gateway Interfaces (mesh NGIs) enabled for communication with one or more shared access networks. Selectively coupling the multiple NGIs to the same shared access network provides redundancy and load balancing aimed at improving the reliability and performance of the network. A first architecture is based on a gateway group, including a plurality of NGIs enabled to communicate with a single shared access network via a designated broadcast server elected from among the NGIs. A second architecture is based on a plurality of (physical) NGIs enabled to communicate with a single shared access network via one or more designated nodes in the shared access network. The designated nodes, or Mesh Servers (MSs), operate as virtual NGIs, and traffic entering or exiting the mesh flows through one of the MSs, thus improving packet broadcast efficiency. |
