In one embodiment, a method includes: receiving, in an edge platform, a plurality of messages from a plurality of edge devices coupled to the edge platform, the plurality of messages comprising metadata including priority information and granularity information; extracting at least the priority information from the plurality of messages; storing the plurality of messages in entries of a pending request queue according to the priority information; selecting a first message stored in the pending request queue for delivery to a destination circuit; and sending a message header for the first message to the destination circuit via at least one interface circuit, the message header including the priority information, and thereafter sending a plurality of packets including payload information of the first message to the destination circuit via the at least one interface circuit. Other embodiments are described and claimed.

Some embodiments of the invention provide a network forwarding element that can be dynamically reconfigured to adjust its data message processing to stay within a desired operating temperature or power consumption range. In some embodiments, the network forwarding element includes (1) a data-plane forwarding circuit (“data plane”) to process data tuples associated with data messages received by the IC, and (2) a control-plane circuit (“control plane”) for configuring the data plane forwarding circuit. The data plane includes several data processing stages to process the data tuples. The data plane also includes an idle-signal injecting circuit that receives from the control plane configuration data that the control plane generates based on the IC's temperature. Based on the received configuration data, the idle-signal injecting circuit generates idle control signals for the data processing stages. Each stage that receives an idle control signal enters an idle state during which the majority of the components of that stage do not perform any operations, which reduces the power consumed and temperature generated by that stage during its idle state.

The invention provides a converged network interface card, a message coding method and a message transmission method thereof. The converged network interface card comprises a PCIE host interface processing module, a high speed network card core logic, a crossbar switch XBAR, an Ethernet network card core logic, an Ethernet message dicing/slicing module, a physical layer, a high speed network/Ethernet message conversion module EoH, and a high speed network/Ethernet configurable network port. The invention supports customized high speed interconnection interface and a standard Ethernet interface on a set of network hardware, and supports three working modes on a set of physical hardware (high speed network mode, Ethernet mode and EoH mode transmitting Ethernet messages over the high speed network), implements seamless compatibility between the high speed network/Ethernet, and flexibly supports multimode applications such as scientific computing and cloud computing.

A system administrator can specify NAT mappings to perform NAT translations in a switch. The administrator can specify an ACL to filter packets to be translated. Filter rules generated from the ACL are stored in a first memory store in a switch and NAT rules generated from the NAT mappings are stored in a second memory store separate from the first memory store. When a packet matches one of the filter rules a tag that identifies the ACL is associated with the packet. When the tagged packet matches one of the NAT rules, the packet is translated according to the matched NAT rule.

Systems and methods for providing multicast group (MCG) membership relative to partition membership in a high performance computing environment. In accordance with an embodiment, by allowing a subnet manager of a local subnet to be instructed that all ports that are members of the relevant partition should be set up as members for a specific multicast group, the SM can perform a more efficient multicast-routing process. It is also possible to limit the IB client interaction with subnet administration conventionally required to handle join and leave operations. Additionally, subnet manager overhead can be reduced by creating a spanning tree for the routing of multicast packets that includes each of the partition members added to the multicast group, instead of creating a spanning tree after each multicast group join request is received, as conventionally required.

Systems and methods for supporting inter subnet control plane protocol for consistent multicast membership and connectivity across multiple subnets in a high performance computing environment. In accordance with an embodiment, by associating a multicast group with an inter-subnet partition, and enforcing a dedicated router port for the multicast group, multicast loop avoidance can be provided for between connected subnets. Because only a single router port is selected as being capable of handling the MC packet, no other router port in the subnet can then pass a multicast packet back to the originating subnet.

A replay device includes: multiple transceiver units that transmits and receives a communication frame, each transceiver unit including a register in which at least data indicating a set-up content relating to an operation of a respective transceiver unit is written; a register access unit that is connected with each transceiver unit through an interface; and a control unit that transmits a control message to the register access unit. The control message includes access target information for designating one or more access target transceiver units, and access content information indicating an access content to a register of each access target transceiver unit. The register access unit sets the one or more target transceiver units designated by the access target information, and perform an access to the register of each access target transceiver unit according to the access content.

A traffic balancing method includes an Ethernet device sending a query packet carrying load information of each physical port of a logical port of the Ethernet device to a controller. The controller receives the query packet and determines a weight factor of each physical port of the logical port based on the load information. The weight factor of each physical port is positively correlated with an amount of traffic that is of the logical port and that can be shared by each of the physical ports. The controller sends a response packet carrying the weight factor to the Ethernet device. The Ethernet device adjusts, based on the weight factor, traffic sent to each of the physical ports.

In one embodiment, a network device includes multiple ports to be connected to a packet data network so as to serve as both ingress and egress ports in receiving and forwarding of data packets including unicast and multicast data packets, a memory coupled to the ports and to contain a combined unicast-multicast user-pool storing the received unicast and multicast data packets, and packet processing logic to compute a combined unicast-multicast user-pool free-space based on counting only once at least some of the multicast packets stored once in the combined unicast-multicast user-pool, compute an occupancy of an egress queue by counting a space used by the data packets of the egress queue in the combined unicast-multicast user-pool, apply an admission policy to a received data packet for entry into the egress queue based on at least the computed occupancy of the egress queue and the computed combined unicast-multicast user-pool free-space.

A routing apparatus includes interconnected switches and a server. At least some of the switches connect to routers belonging to multiple subnetworks, and at least a given switch is configured to receive a packet destined to a subnetwork reachable via the given switch, and forward the packet using forwarding logic of the given switch that supports only a partial subset of the subnetworks, and to monitor traffic flow via the given switch and report traffic flow information indicative of a given subnetwork that is unreachable via the given switch. The server is coupled to the switches, and is configured to receive the traffic flow information from the given switch, and determine switch forwarding information for the given subnetwork, based on the traffic flow information, and to download the switch forwarding information to the given switch, for enabling the forwarding logic to forward subsequent packets destined to the given subnetwork.