A message scheduling method, device and storage medium

Abstract

The embodiment of the present invention discloses a message scheduling method, device, and storage medium. The message scheduling method includes: in the T-th input period, when the first input channel has the first message input, and the second input channel has the first message input When the second message is input, the first authorization code corresponding to when the first input channel schedules the first message is obtained according to the preset acquisition strategy, and the second authorization code corresponding to the second input channel when the second message is scheduled; wherein, T is an integer greater than or equal to 1; when the first authorization code and the second authorization code are the same, detect whether there is a third message input in the first input channel at the T+1 input cycle; when the first input channel is at the T+1 input cycle 1 When there is a third message input in the input cycle, obtain the third authorization code corresponding to the first input channel when scheduling the third message according to the preset acquisition strategy; schedule the first message according to the third authorization code, and obtain the corresponding third authorization code according to the second authorization code The code schedules the second packet.

Description

technical field [0001] The invention relates to network technology, in particular to a message scheduling method, device and storage medium. Background technique [0002] A network processor (Network Processor, NP) is a microprocessor specially applied to a network system. Due to its high speed, low power consumption and programmable advantages, NP has gradually become an irreplaceable part of the routing and forwarding engine. At the same time, the packet processing capability of NP directly affects the performance and quality of network interconnection. After the continuous messages are input to the NP, the NP will evenly dispatch the messages to each microcode processor for parallel processing, wherein the message throughput rate will directly affect the performance and quality of the entire NP. In the prior art, a common message scheduling method is a one-to-many scheduling method, that is, the NP receives a message through an input channel, and then polls and schedules...

AI Technical Summary

Problems solved by technology

[0003] In order to improve the message throughput rate, thereby improving the message scheduling performance of NP, the prior art usually needs to increase the main clock frequency of the chip, but continuously increasing the main clock frequency of the chip will cause the defect of increasing the power consumption of the system, and will also increase the power consumption of the system. The consequences of excessive waste of bandwidth in the NP post-processing module

On the other hand, when the throughput rate of the message is increased by increasing the number of input channels through which the NP receives the message, if the messages of different input channels are scheduled to the same output interface for output, it will cause the defect of message output conflict , can not effectively improve the throughput of the message

Images