Abstract:
This application provides a search apparatus, including a global dispatcher, a global arbiter, and N search engines. The N search engines can access a first search table. The global dispatcher is configured to: determine that a first search keyword is corresponding to the first search table and dispatch the first search keyword to the N search engines. Each search engine is configured to: search, according to a first search algorithm, one subtable to determine whether an entry that matches the first search keyword exists; and output a search result to the global arbiter. The global arbiter is configured to arbitrate the search result output by each search engine, to obtain a search result corresponding to the first search table.
Abstract:
This application discloses a service level adjustment method, apparatus, device, and system, and a storage medium. The method includes: A control device obtains at least one piece of related information in at least one piece of queue status information, a remaining data flow parameter, a current data flow parameter, and a quantity of reporting times of exception information of a target service level of a network device. The control device adjusts, when any information in the related information does not meet a threshold corresponding to the any information, a parameter of the target service level based on a maximum delay associated with the target service level.
Abstract:
A data sending method, where the method includes receiving, by a forwarding device using a first flexible Ethernet (FlexE) group and in multiple timeslots included in a first timeslot set, multiple first encoded data blocks from a physical coding sublayer (PCS), determining, by the forwarding device according to the timeslots included in the first timeslot set and the first FlexE group, a second FlexE group and multiple timeslots included in a second timeslot set, and sending, by the forwarding device, the first encoded data blocks using the second FlexE group and in the timeslots included in the second timeslot set. The forwarding device does not need to process the first encoded data blocks in a conventional layer 2 or layer 3 forwarding mode. Therefore, a processing delay can be reduced, and a transmission delay can be reduced.
Abstract:
Embodiments of the present invention provide a method and an apparatus for generating a decision tree. The method includes generating an encoded rule set according to a rule set, generating a first weighted undirected graph, and calculating a weight of each side in the first weighted undirected graph. If a weight of a side with a maximum weight in the first weighted undirected graph is greater than a first threshold, a first operation is cyclically performed until a weight of a side with a maximum weight in a newly generated weighted undirected graph is less than or equal to the first threshold and a decision tree is generated for a rule subset corresponding to each vertex respectively in the newly generated weighted undirected graph.
Abstract:
Embodiments provide a method, including: receiving, by a forwarding element, a packet through an inbound port; searching for, by the forwarding element, a port table according to the inbound port and determining a first offset, a first length, and an identifier of a first table; determining, by the forwarding element, a first key according to the first offset and the first length, searching for the first table according to the first key, and determining a first instruction; and processing, by the forwarding element, the packet according to the first instruction. In addition, a forwarding element is provided. In the foregoing technical solutions, before processing a packet, the forwarding element does not need to interpret a data format of the packet, so that the forwarding element can flexibly support packets in different data formats.
Abstract:
A method includes receiving, by a forwarding device, M pieces of description information that are sent by a controller and are in a one-to-one correspondence with M forwarding services. Each piece of description information includes preset performance of a corresponding forwarding service and an identifier of a table to be searched, and a set including preset performance of the M forwarding services includes Q elements. The method further includes generating Q pieces of simultaneous access information according to the M pieces of description information; and determining, according to the Q pieces of simultaneous access information, that N tables are to be stored in at least one memory of the forwarding device.
Abstract:
Embodiments provide a method, including: receiving, by a forwarding element, a packet through an inbound port; searching for, by the forwarding element, a port table according to the inbound port and determining a first offset, a first length, and an identifier of a first table; determining, by the forwarding element, a first key according to the first offset and the first length, searching for the first table according to the first key, and determining a first instruction; and processing, by the forwarding element, the packet according to the first instruction. In addition, a forwarding element is provided. In the foregoing technical solutions, before processing a packet, the forwarding element does not need to interpret a data format of the packet, so that the forwarding element can flexibly support packets in different data formats.
Abstract:
This application discloses a method that includes obtaining by a network device an original segment list corresponding to a packet forwarding path, where the original segment list includes original segment identifiers that are sequentially arranged, and each of the original segment identifiers that are sequentially arranged corresponds to one node or link on the packet forwarding path. The method further includes comparing by the network device a plurality of successively adjacent original segment identifiers in the original segment list, to generate a plurality of successively adjacent compressed segment identifiers, where a length of each compressed segment identifier is less than a length of a corresponding original segment identifier, and generating by the network device a compressed segment list, where the compressed segment list includes a plurality of compressed segment identifiers, and the compressed segment list is used to forward a packet along the packet forwarding path.
Abstract:
Embodiments of this application disclose a rule detection method, to increase rule anomaly detection coverage. The method in the embodiments of this application includes: determining, based on an obtained first initial priority corresponding to a first rule, an obtained second current priority corresponding to a second rule, and a determined inclusion relationship between the first rule and the second rule, a first current priority corresponding to the first rule; and then determining, based on a relationship between the first initial priority and the first current priority, whether an anomaly occurs on the first rule.
Abstract:
This application discloses a method that includes obtaining by a network device an original segment list corresponding to a packet forwarding path, where the original segment list includes original segment identifiers that are sequentially arranged, and each of the original segment identifiers that are sequentially arranged corresponds to one node or link on the packet forwarding path. The method further includes comparing by the network device a plurality of successively adjacent original segment identifiers in the original segment list, to generate a plurality of successively adjacent compressed segment identifiers, where a length of each compressed segment identifier is less than a length of a corresponding original segment identifier, and generating by the network device a compressed segment list, where the compressed segment list includes a plurality of compressed segment identifiers, and the compressed segment list is used to forward a packet along the packet forwarding path.