One embodiment of the present invention provides a computing system. The computing system includes a packet processor, a buffer management module, a data monitoring module, and a flow control module. The packet processor identifies a class of service indicating priority-based flow control associated with a remote computing system from a notification message. The buffer management module creates a buffer dedicated for frames belonging to the class of service from the remote computing system in response to identifying the class of service. The data monitoring module detects a potential overflow of the buffer. The flow control module operates in conjunction with the packet processor to generate a pause frame in response to detecting a potential overflow.

Devices and techniques for packet arbitration for buffered packets in a network device are described herein. A packet can be received at an input of the network device. The packet can be placed in a buffer for the input and a characteristic of the packet can be obtained. A record for the packet, that includes the characteristic, is written into a data structure that is independent of the buffer. Arbitration, based on the characteristic of the packet in the record, can then be performed among multiple packets to select a next packet from the buffer for delivery to an output.

The present invention relates to a method of managing congestion in a multi-path network, the network having a plurality of network elements arranged in a plurality of switch stages and a plurality of network links interconnecting the network elements, the method comprising the steps of detecting congestion on a network link, the congested network link interconnecting the output port of a first network element with a first input port of a second network element in a subsequent switch stage; identifying an uncongested network link connected to a second input port of said second network element; and directing future data packets on a route across the multi-path network which includes the identified uncongested network link. Also provided is a multi-path network and an Ethernet bridge or router incorporating such a multi-path network.

A method, apparatus, and system for implementing node port virtualization on a fiber channel in the field of communication technologies are provided. Multiple different virtual Node Port (N_port) identifications (IDs) are allocated to each of multiple N_ports, of an N_port virtualization (NPV) switch, corresponding to an N_port ID of a remote node. Because a virtual N_port ID is allocated, to each remote node, for more than one N_port of the NPV switch, regardless of which N_port of these N_ports a node connected to the NPV switch is registered for, the node can obtain the virtual N_port ID that is of each remote node and that is corresponding to the N_port. Therefore, any node connected to the NPV switch can communicate with any remote node, thereby improving communication efficiency.

A plurality of management cards including an active card and a standby card are provided. The active card determines open or block of a ring port in accordance with an event based on a ring protocol, issues an open instruction or a block instruction to a line card, and notifies a block factor in addition to the block instruction when issuing the block instruction. The line card controls open or block of the ring port in accordance with the open instruction or the block instruction and retains open/block information of the ring port and a block factor of the block state in a port management table. When the standby card is changed to the active card in accordance with a predetermined change instruction, it acquires the information retained in the port management table from the line card.

Devices and methods for policing traffic flows in a network are described herein. For example, a network device can include a processing unit and a memory operably coupled to the processing unit. The network device can be configured to provide an input port and an output port, the input and output ports being configured to process a traffic flow received at the network device. The network device can also be configured to queue the traffic flow using a plurality of buffers associated with the output port, cyclically direct the traffic flow from the input port to each of the plurality of buffers according to a queuing schedule, and enforce a policer policy synchronized to the queuing schedule. The policer policy can ensure that the traffic flow does not exceed a predetermined bandwidth.

Disclosed are various embodiments for multi-homing in an extended bridge, including both multi-homing of port extenders and multi-homing of end stations. In various embodiments, a controlling bridge device receives a packet via an ingress virtual port and determines a destination virtual port link aggregation group based at least in part on a destination media access control (MAC) address of an end station in the packet. The controlling bridge device selects one of multiple egress virtual ports of the destination virtual port link aggregation group. The end station of the extended bridge is reachable through any of the egress virtual ports of the destination virtual port link aggregation group. The controlling bridge device forwards the packet through the selected egress virtual port, and the forwarded packet includes an identifier of a destination virtual port to which the end station is connected.

System and method for supporting flexible forwarding domain boundaries in a high performance computing environment. In accordance with an embodiment, flexible forwarding domain boundaries can be supported by dividing/partitioning a physical switch into two or more logical switches, where each switch is logically in a different domain, and allowing a fabric to be decomposed into independent subnets with one two or more physical end ports at the physical switch. By doing so, the same hierarchical forwarding structure and management structure between subnets can be provided as when complete physical switches are used as building blocks.

System and method for aggressive credit waiting in a high performance computing environment. In accordance with an embodiment, systems and methods can provide for an indexed matrix of credit wait policies between ports within a single switch. In addition, systems and methods can provide for an array of credit wait polices at an egress port from a switch, the array being indexed by virtual lane.

A method for connecting endpoints. The method may include determining, using a first group of software threads, cross-connection information regarding various cross-connections among various network elements. The network elements may include a first network element and a second network element. The method may further include determining, using the first group of software threads, link information regarding various links between the network elements. The method may further include processing, within a network cache, the cross-connection information and the link information into various network resources groups. Each network resource group of the network resources groups may describe adjacent cross-connections for a link among the links.