Systems and methods for providing input and output ports to connect to channels are provided. Input and output ports are the basic building blocks to create more complex data routing IP blocks. By aggregating these modular ports in different ways, different implementations of crossbar or Network on Chip (NoC) can be implemented, allowing flexible routing structure while maintaining all the benefits of channels such as robustness against delay variation, data compression and simplified timing assumptions.

A receiving core reads a packet from an ingress interface, wherein the ingress interface is an interface corresponding to a forwarding group to which the receiving core belongs; the receiving core sends the read packet to a forwarding core in the forwarding group; and the forwarding core sends the packet to a corresponding egress interface.

Systems and methods for providing input and output ports to connect to channels are provided. Input and output ports are the basic building blocks to create more complex data routing IP blocks. By aggregating these modular ports in different ways, different implementations of crossbar or Network on Chip (NoC) can be implemented, allowing flexible routing structure while maintaining all the benefits of channels such as robustness against delay variation, data compression and simplified timing assumptions.

A multi-port queue group system an Network Processing Unit coupled to ingress port(s) and an egress port group having a first egress port and a second egress port. The NPU includes an egress queue group having a first egress queue associated with the first egress port and a second egress queue associated with the second egress port. The NPU receives data packets that are each directed to the egress port group via the ingress port(s), and buffers a first subset of the data packets in the first egress queue included in the egress queue group, and a second subset of the data packets in the second egress queue included in the egress queue group. The NPU then transmits at least one of the data packets via at least one of the first egress port and the second egress port included in the egress port group.

An inter-cloud communication method, used to implement communication between two clouds, where virtual machines belonging to a same virtual network are created in the two clouds. A receive end cloud uses a gateway node as an entrance to external communication, and all data packets to be sent to a virtual machine in the receive end cloud are sent to the gateway node, thereby preventing a location change of the virtual machine from affecting a transmit end cloud. In addition, the data packet only needs to pass through the gateway node in the receive end cloud and a computing node on which the virtual machine that receives the data packet is located, that is, the data packet only needs two hops to reach a destination, thereby shortening a communication path, and improving inter-cloud communication efficiency.

A technique includes receiving a packet at a network device, wherein the packet is to be routed in a network to a destination network device; determining a plurality of candidate routes for the packet to be routed to the destination network device; grouping the plurality of candidate routes into a first set of candidate routes and a second set of candidate routes based on hop counts associated with the plurality of candidate routes; selecting one of the first or second sets based on a congestion metric threshold; selecting a candidate route from the selected first or second set based on weight metric values associated with the candidate routes of the selected first or second set; and selecting an egress port associated with the selected candidate route.

Systems and methods for providing input and output ports to connect to channels are provided. Input and output ports are the basic building blocks to create more complex data routing IP blocks. By aggregating these modular ports in different ways, different implementations of crossbar or Network on Chip (NoC) can be implemented, allowing flexible routing structure while maintaining all the benefits of channels such as robustness against delay variation, data compression and simplified timing assumptions.

A multi-port queue group system an Network Processing Unit coupled to ingress port(s) and an egress port group having a first egress port and a second egress port. The NPU includes an egress queue group having a first egress queue associated with the first egress port and a second egress queue associated with the second egress port. The NPU receives data packets that are each directed to the egress port group via the ingress port(s), and buffers a first subset of the data packets in the first egress queue included in the egress queue group, and a second subset of the data packets in the second egress queue included in the egress queue group. The NPU then transmits at least one of the data packets via at least one of the first egress port and the second egress port included in the egress port group.

Systems and methods for providing input and output ports to connect to channels are provided. Input and output ports are the basic building blocks to create more complex data routing IP blocks. By aggregating these modular ports in different ways, different implementations of crossbar or Network on Chip (NoC) can be implemented, allowing flexible routing structure while maintaining all the benefits of channels such as robustness against delay variation, data compression and simplified timing assumptions.

A technique includes receiving a packet at a network device of a plurality of network devices in a network. The packet is to be routed to a destination network device. The technique includes determining, by a routing engine of the network device, a plurality of candidate routes for the packet to be routed to the destination network device; and grouping, by the routing engine, the plurality of candidate routes into a first set of candidate routes and a second set of candidate routes based on hop counts that are associated with the plurality of candidate routes. The technique includes the routing engine selecting one of the first or second sets based on a congestion metric threshold; and the routing engine selecting a candidate route from the selected first or second set based on weight metric values that are associated with the candidate routes of the selected first or second set. The technique includes the routing engine selecting an egress port associated with the selected candidate route.