Packet forwarding

1. A method for forwarding a packet, applicable to each non-uniform memory architecture (NUMA) node included in a network function virtualization (NFV) device, wherein the each NUMA node comprises a forwarding group comprising a receiving core and a forwarding core, at least one core of a plurality of cores included in the each NUMA node and at least one core of the other cores of the plurality of cores being configured as a receiving core and a forwarding core of the forwarding group of the each NUMA node respectively, and the method comprises:

creating, by a designated core in the NFV device, the forwarding group for the NUMA node, wherein the forwarding group comprises all cores of the NUMA node;

configuring, by the designated core, the cores in the forwarding group as a control core, a receiving core or a forwarding core;

configuring, by the designated core, a corresponding ingress interface for the forwarding group;

reading, by a receiving core of the each NUMA node, a packet from an ingress interface, wherein the ingress interface is an interface corresponding to a forwarding group to which the receiving core of the each NUMA node belongs;

sending, by the receiving core of the each NUMA node, the packet read by itself to a forwarding core in the forwarding group to which the receiving core of the each NUMA node belongs; and

sending, by the forwarding core in the forwarding group to which the receiving core of the each NUMA node belongs, the packet sent by the receiving core of the each NUMA node to a corresponding egress interface.

2. The method according to claim 1 , wherein configuring the cores in the forwarding group as a control core, a receiving core or a forwarding core comprises:

configuring, by the designated core, the core itself as a control core;

in case that the forwarding group does not comprise the control core,

configuring, by the control core, at least one core in the forwarding group as the receiving core according to a predesigned rule, and

configuring, by the control core, the remaining cores in the forwarding group as the forwarding cores; and

in case that the forwarding group comprises the control core,

configuring, by the control core, at least one core in the forwarding group except the control core itself as the receiving core according to the predesigned rule, and

configuring, by the control core, the remaining cores in the forwarding group as the forwarding cores.

3. The method according to claim 2 , wherein sending the read packet to a forwarding core in the forwarding group comprises:

acquiring, by the receiving core, a corresponding relationship among a receiving core, a forwarding core and a forwarding group;

determining, by the receiving core, a forwarding core in the forwarding group according to the corresponding relationship; and

sending, by the receiving core, the read packet to the determined forwarding core.

4. The method according to claim 2 , wherein configuring a corresponding ingress interface for the forwarding group comprises:

identifying, by the control core, a network interface card associated with the NUMA node, and

allocating, by the control core, an interface of the network interface card as the ingress interface, to the forwarding group of the NUMA node.

5. The method according to claim 2 , wherein configuring a corresponding ingress interface for the forwarding group comprises:

allocating, by the control core, a specified interface to the forwarding group as the ingress interface according to an interface configuration command.

6. A device for forwarding a packet, comprising:

a receiving core;

a forwarding core;

a control core which is a designated core of the device for forwarding a packet; and

a non-transitory machine readable storage medium storing machine readable instructions,

wherein the receiving core and the forwarding core which belong to a same forwarding group included in each non-uniform memory architecture (NUMA) node, the each NUMA node comprising a plurality of cores, at least one core of the plurality of cores included in the each NUMA node and at least one core of the other cores of the plurality of cores being configured as a receiving core and a forwarding core of the forwarding group of the each NUMA node respectively,

the machine readable instructions cause the control core to:

create the forwarding group for the NUMA node, wherein the forwarding group comprises all cores of the NUMA node;

configure the cores in the forwarding group as a control core, a receiving core or a forwarding core; and

configure a corresponding ingress interface for the forwarding group,

the machine readable instructions are executable by the receiving core of the each NUMA node to:

read a packet from an ingress interface, wherein the ingress interface is an interface corresponding to a forwarding group to which the receiving core of the each NUMA node belongs;

send the read packet to a forwarding core in the forwarding group to which the receiving core of the each NUMA node belongs; and

the machine readable instructions are executable by the forwarding core in the forwarding group to which the receiving core of the each NUMA node belongs to:

send the packet sent by the receiving core of the each NUMA node to a corresponding egress interface.

7. The device according to claim 6 , wherein when configuring the cores in the forwarding group as a control core, a receiving core or a forwarding core, the machine readable instructions further cause the control core to:

configure itself as a control core;

in case that the forwarding group does not comprise the control core,

configure at least one core in the forwarding group as the receiving core according to a predesigned rule, and

configure the remaining cores in the forwarding group as the forwarding cores; and

in case that the forwarding group comprises the control core,

configure at least one core in the forwarding group except the control core itself as the receiving core according to the predesigned rule, and

configure the remaining cores in the forwarding group as the forwarding cores.

8. The device according to claim 7 , wherein when sending the read packet to a forwarding core in the forwarding group, the machine readable instructions further cause the receiving core to:

acquire a corresponding relationship among a receiving core, a forwarding core and a forwarding group;

determine a forwarding core in the forwarding group according to the corresponding relationship; and

send the read packet to the determined forwarding core.

9. The device according to claim 7 , wherein when configuring a corresponding ingress interface for the forwarding group, the machine readable instructions further cause the control core to:

identify a network interface card associated with the NUMA node, and

allocate an interface of the network interface card as the ingress interface, to the forwarding group of the NUMA node.

10. The device according to claim 7 , wherein when configuring a corresponding ingress interface for the forwarding group, the machine readable instructions further cause the control core to:

allocate a specified interface to the forwarding group as the ingress interface according to an interface configuration command.