摘要:
A method of rate-control for a group of ports in a communication network. Ramping is performed for allowed-rate of each of the group of ports, if sum of the ingress-rates of the group of ports is less than a given rate threshold; and adjusting is performed for allowed-rate of each of the group of ports, if sum of the ingress-rates of the group of ports is greater than or equal to the given rate threshold, such that the sum of allowed-rates of the group of ports is equal to the given rate threshold.
摘要:
A method of rate-control for a group of ports in a communication network. Ramping is performed for allowed-rate of each of the group of ports, if sum of the ingress-rates of the group of ports is less than a given rate threshold; and adjusting is performed for allowed-rate of each of the group of ports, if sum of the ingress-rates of the group of ports is greater than or equal to the given rate threshold, such that the sum of allowed-rates of the group of ports is equal to the given rate threshold.
摘要:
An apparatus comprising an edge virtual bridging (EVB) bridge, and an EVB station coupled to the EVB bridge, wherein the EVB station is configured to send to the EVB bridge a virtual station interface (VSI) discovery and configuration protocol (VDP) request comprising a filter information (info) field without specifying a virtual local area network (VLAN) identifier (ID), and wherein the EVB bridge is configured to send a VLAN ID (VID) to the EVB station in a second filter info field in a VDP response to the VDP request.
摘要:
A network comprising a switch configured to learn a source address (SA) for a source node with a source port that receives a frame comprising the SA if the SA is not previously learned on another port and to subsequently forward any frame comprising a destination address (DA) for a destination node that matches the learned SA on the source port associated with the SA. A network comprising a switch configured to flood a frame received on a receiving port if the frame comprises a DA that is not previously learned on a source port and to distinguish the flooded frame from other forwarded frames by labeling the flooded frame, wherein the frame is flooded on a plurality of ports. A method comprising disabling a plurality of filtering database (FDB) entries that correspond to a plurality of addresses that are associated with a port, and sending the addresses in a control frame.
摘要:
An apparatus comprising a processor configured to implement a method comprising receiving a first data structure comprising a first virtual local area network identifier (VID) and a first forwarding type, comparing the first VID and a first forwarding type with a second VID and a second forwarding type in a second data structure, and reporting a misprovisioned data structure if the first VID is the same as the second VID and the first forwarding type is not the same as the second forwarding type. Also disclosed is a method comprising implementing a network protocol to forward provisioning information in a plurality of VID Tables corresponding to a plurality of nodes, and analyzing the provisioning information to detect misprovisioned nodes.
摘要:
A communications network, comprising a plurality of nodes supporting connection-oriented traffic and connectionless traffic, wherein management traffic between the nodes is propagated as connectionless traffic having a unique management traffic Virtual Local Area Network (VLAN) identifier (VID), and wherein at least one node is configured to perform a Link Layer Discovery Protocol (LLDP) exchange operation to enable detection of inconsistent provisioning of the Management VID in at least one node. Also disclosed is a communications network component, comprising logic that enables connection-oriented traffic and VLAN-based connectionless traffic, wherein the logic propagates management messages as VLAN-based connectionless traffic having a management VID, and wherein the logic enables detection of inconsistent management VIDs in a network.
摘要:
An apparatus comprising a service network and a plurality of Layer 2 networks at a plurality of different physical locations coupled to the service network via a plurality of edge nodes at the Layer 2 networks, wherein the edge nodes are configured to maintain a plurality of Internet Protocol (IP) addresses of a plurality of hosts across the Layer 2 networks, and wherein the IP addresses of the hosts in each of the Layer 2 networks are mapped by the other Layer 2 networks to a Media Access Control (MAC) address of each of the edge nodes in the same Layer 2 networks of the hosts.
摘要:
A method comprising receiving a data frame, encapsulating the received data frame into a reflected frame message (RFM) encapsulated frame, and if the received data frame causes the RFM encapsulated frame to exceed a maximum service data unit size, either the data frame is truncated or the data frame is split into two smaller frames and those two smaller frames are encapsulated by two separate RFM frames, wherein if a truncation flag is set to be true, the data frame is truncated to be encapsulated in one RFM.
摘要:
A data communications network includes a ring configured for spatial reuse, such as a resilient packet ring. A first bridge is coupled to the ring and to an end station such as an interworking bridge or router. A second bridge coupled to the ring learns an association between the first bridge and the end station by monitoring packets originated in the network by the end station and sent on the ring by the first bridge. The second bridge learns the association by obtaining an identifier of the first bridge and an address of the end station from received packets. When the second bridge receives packets destined for the end station, the packets are forwarded as broadcast transmissions on the ring if the association between the first bridge and the end station has not yet been learned, and are forwarded as unicast transmissions to the first bridge on the ring if the association between the first bridge and the end station has been learned.
摘要:
This invention defines a routing method which allows a determination of best paths from link weights in situations where these weights are not additive, but the characteristic of a concatenated path is determined mainly by single "bottleneck" links. To be compatible, the metric must be chosen from a specific set of metrics, called bottleneck metrics in this description. The new widest-path method has some characteristics in common with the known Dijkstra method for additive weights. Implementations in communication networks are disclosed which lead to new nodes, new networks and new network protocols.