Abstract: A multi-thread systolic array includes a plurality of processing elements, each including a processor. Each of the processing elements is configured to: receive a plurality of first inputs from a respective first input source; receive a plurality of second inputs from a respective second input source; the plurality of first inputs and the plurality of second inputs being arranged as a plurality of pairs corresponding to a plurality of threads; schedule, for each operation cycle of the processor, a certain thread of the plurality of threads; and execute a computation operation for the certain thread.

Abstract: A multi-thread systolic array includes a plurality of processing elements, each including a processor. Each of the processing elements is configured to: receive a plurality of first inputs from a respective first input source; receive a plurality of second inputs from a respective second input source; the plurality of first inputs and the plurality of second inputs being arranged as a plurality of pairs corresponding to a plurality of threads; schedule, for each operation cycle of the processor, a certain thread of the plurality of threads; and execute a computation operation for the certain thread.

Abstract: A data packet forwarding method and a network device, where the method includes, receiving a data packet, obtaining a local physical address and a destination physical address of the data packet when the entry matching the data packet does not exist in the first match table, determining whether the local physical address is the same as the destination physical address, searching a second match table for an entry in which a physical address matches the destination physical address when the local physical address is different from the destination physical address, generating a layer 2 forwarding entry according to the entry matching the destination physical address, and forwarding the data packet according to the layer 2 forwarding entry. Therefore, processing efficiency of the new data packet and network performance are improved.

Abstract: An operations, administration, and maintenance (OAM) packet processing method, a network device, and a network system are provided. The method includes receiving a first Internet Protocol (IP) packet, where the first IP packet includes an OAM packet, an IP packet header of the OAM packet includes a first source device identifier (ID) and a first sink device ID, and returning a second IP packet to reply to the OAM packet, where the second IP packet includes an OAM reply packet, an IP packet header of the OAM reply packet includes a second source device ID, a second sink device ID, and a second packet type, the second source device ID is an ID of a local network device, and the second sink device ID is the same as the first source device ID. An IP packet header may be used to carry OAM information, which improves a fault detection capability in an OAM method.

Abstract: A data packet forwarding method and a network device, where the method includes, receiving a data packet, obtaining a local physical address and a destination physical address of the data packet when the entry matching the data packet does not exist in the first match table, determining whether the local physical address is the same as the destination physical address, searching a second match table for an entry in which a physical address matches the destination physical address when the local physical address is different from the destination physical address, generating a layer 2 forwarding entry according to the entry matching the destination physical address, and forwarding the data packet according to the layer 2 forwarding entry. Therefore, processing efficiency of the new data packet and network performance are improved.

Abstract: An operations, administration, and maintenance (OAM) packet processing method, a network device, and a network system are provided. The method includes receiving a first Internet Protocol (IP) packet, where the first IP packet includes an OAM packet, an IP packet header of the OAM packet includes a first source device identifier (ID) and a first sink device ID, and returning a second IP packet to reply to the OAM packet, where the second IP packet includes an OAM reply packet, an IP packet header of the OAM reply packet includes a second source device ID, a second sink device ID, and a second packet type, the second source device ID is an ID of a local network device, and the second sink device ID is the same as the first source device ID. An IP packet header may be used to carry OAM information, which improves a fault detection capability in an OAM method.