摘要:
Embodiments of this application provide a packet processing method and apparatus, a communications device, and a switching circuit. The method includes: receiving, by a first device, a packet from a second device; determining, by the first device, a first queue buffer used to store the packet; determining, by the first device, a first upper limit value of the first queue buffer based on an available value of a first port buffer and an available value of a global buffer, where the global buffer includes at least one port buffer, the first port buffer is one of the at least one port buffer, the first port buffer includes at least one queue buffer, and the first queue buffer is one of the at least one queue buffer; and processing, by the first device, the packet based on the first upper limit value of the first queue buffer, an occupation value of the first queue buffer, and a size of the packet. In the embodiments of this application, differentiated packet discarding is implemented.
摘要:
This application provides a packet forwarding method, a forwarding device, and a network device. The method includes: receiving N Time-Sensitive Networking TSN packet flows, where each of the N TSN packet flows is corresponding to a constraint condition that defines duration of a single cycle, a maximum quantity of packets that are allowed to be transmitted in a single cycle, and a maximum length of a single packet; and forwarding the N TSN packet flows based on a new constraint condition that defines duration of a single cycle, a maximum quantity of packets that are allowed to be transmitted in a single cycle, and a maximum length of a single packet, where each TSN packet flow is forwarded in a case in which a corresponding constraint condition is complied with, and the new constraint condition is determined based on the constraint condition corresponding to each of the N TSN packet flows. The method in this application helps reduce jitter of a packet during forwarding, and can improve resource utilization and forwarding efficiency of forwarding the packet by a system.
摘要:
A data processing method executed by a network apparatus, and a related device are provided. The data processing method executed by a network apparatus includes: receiving, by the i th processing circuit in a first circuit set, a first packet header and data D (1, i-1) , obtaining data D' (1, i) based on a first field in the first packet header, and sending the first packet header and data D (1, i) to the (i+1) th processing circuit in the first circuit set, where the data D (1, i) is obtained based on the data D (1, i-1) and the data D' (1, i) ; sending, by the i th processing circuit in the first circuit set, the data D (1, i) to the (i+1) th processing circuit in a second circuit set; and sending, by the i th processing circuit in the second circuit set, a second packet header to the (i+1) th processing circuit in the second circuit set. The solutions provided in embodiments of the present invention can help to increase a utilization rate of a processing resource and reduce packet processing complexity.
摘要:
This application provides a packet forwarding method, a forwarding device, and a network device. The method includes: receiving N Time-Sensitive Networking TSN packet flows, where each of the N TSN packet flows is corresponding to a constraint condition that defines duration of a single cycle, a maximum quantity of packets that are allowed to be transmitted in a single cycle, and a maximum length of a single packet; and forwarding the N TSN packet flows based on a new constraint condition that defines duration of a single cycle, a maximum quantity of packets that are allowed to be transmitted in a single cycle, and a maximum length of a single packet, where each TSN packet flow is forwarded in a case in which a corresponding constraint condition is complied with, and the new constraint condition is determined based on the constraint condition corresponding to each of the N TSN packet flows. The method in this application helps reduce jitter of a packet during forwarding, and can improve resource utilization and forwarding efficiency of forwarding the packet by a system.
摘要:
Embodiments of this application relate to the field of communication, and disclose a packet processing method and apparatus, to alleviate a problem of high power consumption of a processor in a pipeline architecture in the conventional technology. A specific solution is as follows: A packet parsing module parses a first packet, and extracts first packet information from the first packet if it is determined that service complexity corresponding to the first packet is lower than a preset degree. A data amount of the first packet information is less than a data amount of second packet information, the second packet information is information extracted from a second packet, and service complexity of the second packet is higher than the preset degree. The packet parsing module sends the first packet information to a packet distribution module. The packet distribution module receives the first packet information, and sends the first packet information to a processor pipeline module. The processor pipeline module receives the first packet information, and processes the first packet information.
摘要:
A method, device, and system for implementing load balancing are provided, and are applied to a 4-node network structure. Every two nodes in the 4-node network structure are interconnected, and the nodes are, for example, dies. The 4-node network structure includes a source node and a destination node. The method includes: When a bandwidth occupied by ingress traffic flowing into the source node and destined for the destination node is greater than a bandwidth of a fabric side link between the source node and the destination node, the source node selects at least two transmission paths to send the ingress traffic to the destination node; and when the bandwidth occupied by the ingress traffic is less than or equal to the bandwidth of the fabric side link between the source node and the destination node, the source node transmits the ingress traffic on a direct link between the source node and the destination node. This method can save network resources while implementing non-blocking switching of traffic, and meet a low-network delay requirement to some extent.
摘要:
A data processing method executed by a network apparatus, and a related device are provided. The data processing method executed by a network apparatus includes: receiving, by the i th processing circuit in a first circuit set, a first packet header and data D (1, i-1) , obtaining data D' (1, i) based on a first field in the first packet header, and sending the first packet header and data D (1, i) to the (i+1) th processing circuit in the first circuit set, where the data D (1, i) is obtained based on the data D (1, i-1) and the data D' (1, i) ; sending, by the i th processing circuit in the first circuit set, the data D (1, i) to the (i+1) th processing circuit in a second circuit set; and sending, by the i th processing circuit in the second circuit set, a second packet header to the (i+1) th processing circuit in the second circuit set. The solutions provided in embodiments of the present invention can help to increase a utilization rate of a processing resource and reduce packet processing complexity.
摘要:
This application discloses a communications method and a related device. The network device includes a first node and a second node, and the method includes: the first node obtains a first target bandwidth of first traffic on a first path, where the first path is a path used to transmit the first traffic from the first node to the second node, the first target bandwidth is not greater than a minimum value in remaining available bandwidths of single-hop links on the first path, and a single-hop link is a link between two directly connected nodes. The first node sends the first traffic on the first path based on the first target bandwidth. In this way, before traffic is transmitted in the network device, available bandwidth resources are predetermined for the traffic, and the traffic corresponding to the available bandwidth resources is transmitted on the determined available bandwidth resources. This ensures that no congestion occurs during traffic transmission on a path, and improves effectiveness and reliability of traffic transmission between nodes in the network device.
摘要:
A packet edit processing method and a related device. The packet edit processing method includes: generating an input packet template based on N - to-be-edited protocol header fields of an input packet, and a preset protocol field mapping relationship, where the input packet template includes N protocol descriptors that are corresponding, in a one-to-one manner, to the N protocol header fields, arid the protocol field mapping relationship is a mapping relationship between a protocol field included in a protocol descriptor and a protocol field included in a protocol header field that has a correspondence with the protocol descriptor; performing edit processing on the input packet template to obtain an output packet template; and converting, based on the preset protocol field mapping relationship, M protocol descriptors included in the obtained output packet template into M protocol header fields of an output packet, and replacing the N protocol header fields in the input packet with the M protocol header fields to obtain the output packet. The foregoing solution helps reduce instruction complexity for packet edit processing and improve efficiency of packet edit processing.