An example system for driver-to-driver communication can include a first driver located on a first network device and including a transmit data mover (XDM) to send a preformatted message over a fabric interconnect to a second driver located on a second network device. The example system can also include the second driver located on the second network device and including a receive data mover (RDM) to receive the preformatted message, generate an interrupt responsive to receipt of the preformatted message, and route the interrupt to the second driver. The second driver can read the preformatted message responsive to receipt of the interrupt.

An interconnect is provided that has a plurality of nodes, and a ring network to which each of the nodes is connected to allow packets to be transmitted between nodes. For an ordered sequence of packets one of the nodes is arranged as a source node to add each packet of the ordered sequence on to the ring network, and another of the nodes is arranged as a destination node to remove each packet of the ordered sequence from the ring network. The source node is enabled to add a packet of the ordered sequence on to the ring network without waiting for a previously added packet of the ordered sequence to be removed from the ring network by the destination node. When the destination node is unable to accept a given packet of the ordered sequence that is currently being presented to the destination node by the ring network, that given packet remains on the ring network and continues to be transmitted around the ring network such that after a respin period that given packet will be presented again to the destination node. The destination node is then arranged to prevent acceptance of at least any other packets of the ordered sequence subsequently presented to the destination node by the ring network until the destination node has accepted the given packet following at least one respin period. This can improve the efficiency of the ring network in the handling of ordered sequences of packets, whilst still ensuring the ordering constraints are met.

A traffic manager is shared amongst two or more egress blocks of a network device, thereby allowing traffic management resources to be shared between the egress blocks. Among other aspects, this may reduce power demands and allow a larger amount of buffer memory to be available to a given egress block that may be experiencing high traffic loads. Optionally, the shared traffic manager may be leveraged to reduce the resources required to handle data units on ingress. Rather than buffer the entire unit in the ingress buffers, an arbiter may be configured to buffer only the control portion of the data unit. The payload of the data unit, by contrast, is forwarded directly to the shared traffic manager, where it is placed in the egress buffers. Because the payload is not being buffered in the ingress buffers, the ingress buffer memory may be greatly reduced.

An automated Link Aggregation Group (LAG) configuration system includes a plurality of slave switch devices that are each coupled to an endhost device by at least one respective link. Each of the plurality of slave switch devices receives a Link Aggregation Group (LAG) communication from the endhost device, and forwards endhost device information in that LAG communication to a master switch device. The master switch device receives endhost device information from each of the plurality of slave switch devices and determines that each of the plurality of slave switch devices are coupled to the endhost device. In response, the master switch device sends a LAG instruction to each of the plurality of slave switch devices that causes the at least one respective link that couples each of the plurality of slave switch devices to the endhost device to be configured in a LAG.

A switching fabric configuration and management system includes switch devices. Each of the switch devices identifies a domain in a switching fabric database in that switch device, and determines others of the switch devices that share the domain. Each of the switch devices may then communicate with the others of the switch devices that share the domain to designate a master switch device and one or more slave switch devices from the switch devices that share the domain. The designation of the master switch device and the one or more slave switch devices configures the master switch device and the one or more slave switch devices as a switching fabric. Each slave switch devices then reports their slave switch device status to the master switch device, and the master switch device performs at least one control operation on the at least one slave switch device.

An example system for driver-to-driver communication can include a first driver located on a first network device and including a transmit data mover (XDM) to send a preformatted message over a fabric interconnect to a second driver located on a second network device. The example system can also include the second driver located on the second network device and including a receive data mover (RDM) to receive the preformatted message, generate an interrupt responsive to receipt of the preformatted message, and route the interrupt to the second driver. The second driver can read the preformatted message responsive to receipt of the interrupt.

A ring interconnect system comprises a plurality of nodes. Each node is connected to two other nodes to form a ring interconnect. Every pair of nodes is connected by an inter-node path for that pair of nodes distinct from the ring interconnect. Each of the nodes comprises a message buffer to buffer messages received from at least one device associated with the node. Each of the nodes also comprises activity level circuitry to transmit an activity indication, when a number of the messages in the message buffer is equal to or above a threshold, to each other node of the plurality of nodes via the respective inter-node paths. Each of the nodes also comprises arbitrator circuitry to receive the activity indications from each other node and from the activity level circuitry, and to allow ingress of a message from the message buffer onto the ring interconnect in dependence on the activity indications. Also provided is a method of operating a node of a ring interconnect system

A data bus subscriber and a method for processing data, wherein the data bus subscriber can be connected to a local bus, particularly a ring bus, and the data bus subscriber has an input interface, which can be connected to the local bus, for receiving first local bus data, an output interface, which can be connected to the local bus, for transmitting second local bus data, a processing component for synchronous processing of the first local bus data and/or data stored in a memory and for output of at least one control signal, a logic unit, which is adapted in order to modify a quantity of received first local bus data based on the control signal in order to generate the second local bus data to be transmitted, wherein the logic unit is further adapted for synchronous, delayed transmitting of the second local bus data via the output interface.

Significantly optimized multi-stage networks with scheduling methods for faster scheduling of connections, useful in wide target applications, with VLSI layouts using only horizontal and vertical links to route large scale sub-integrated circuit blocks having inlet and outlet links, and laid out in an integrated circuit device in a two-dimensional grid arrangement of blocks are presented. The optimized multi-stage networks in each block employ several slices of rings of stages of switches with inlet and outlet links of sub-integrated circuit blocks connecting to rings from either left-hand side only, or from right-hand side only, or from both left-hand side and right-hand side; and employ multi-drop links where outlet links of cross links from switches in a stage of a ring in one sub-integrated circuit block are connected to either inlet links of switches in the another stage of a ring in the same or another sub-integrated circuit block.

Techniques are described for detecting egress network devices of a point-to-multipoint (P2MP) label switched path (LSP). For example, a network device may include one or more processors configured to identify a P2MP LSP for receiving multicast traffic from a multicast source for a specific multicast group for which the network device has an interested receiver, wherein the network device is to be an egress network device of the P2MP LSP; and send, to an ingress network device of the P2MP LSP, a P2MP egress identification message to add the network device as an egress network device of the P2MP LSP, wherein the one or more processors are further configured to output the P2MP egress identification message into a multipoint-to-point (MP2P) ring LSP for which the ingress network device of the P2MP LSP is a sole egress network device of the MP2P ring LSP.