Packet scheduling method and device

Abstract

The invention discloses a packet scheduling method and a device. The method comprises: A. a received packet is cached into corresponding VOQ according to service levels and destination contained in a packet head; B. according to the number of the packet in each VOQ and network congestion level, authorization is generated for each VOQ and is sent to each VOQ; C. packet scheduling dequeuing is carried out by each VOQ according to the authorization; D. round-robin scheduling is performed according to the priority level of dequeuing scheduling, and then, packet scheduling is output. The invention simultaneously discloses a device for realizing packet scheduling, and the method and the device can realize packet scheduling to VOQ, are flexible to schedule and save hardware logics resources.

Description

technical field [0001] The present invention relates to network packet transmission technology, in particular to a packet scheduling method and device. Background technique [0002] With the continuous development of communication technologies, operators need to meet all-round service requirements of users. From the integration of telecommunication network, radio and television network and mobile network, as well as the current domestic telecommunication restructuring, it can be seen that it is the general trend to enable operators to have full-service operation capabilities through the integration of various services. This requires that a communication network must support multiple services, but each service has different quality of service (QoS) requirements. QoS refers to a series of service requests that the network needs to meet when transmitting data streams and the mechanism to realize these service requests. These service requests can be measured by a series of indi...

AI Technical Summary

Problems solved by technology

But because the scheduling level is already fixed, the flexibility is low

At the same time, this scheduling method can only be scheduled after the traffic reaches the scheduling node, and the scheduling on the input side fails to consider the degree of network congestion. Therefore, all the traffic on the input side will be scheduled and transmitted to the output side for buffering, waiting for the output side. Scheduling output, which may not only lead to congestion in the scheduling of the output side, but also cause problems in the cooperation between the input side and the output side, and greatly consume the hardware logic resources of the output side

In addition, since the scheduling level is fixed and not configurable, the support for various standards is limited, and only limited quality of service can be guaranteed

Images