Message speed limiting method based on token barrel

Abstract

This invention provides one message limit method based on order barrel, which comprises the following steps: receiving certain message data triggering one filling order operation and filling certain order into barrel; judging whether order is over message section length and if order number is larger or equal to data section length then permitting data section and minus message data length; if order length is less than message length then discarding data section not repeating judge and to restore the data section to the order barrel.

Description

technical field [0001] The invention relates to a method for limiting the message rate in the data network field, in particular to a hardware implementation method for limiting the message rate by using a token bucket in the data network. Background technique [0002] In an existing data network, due to the burstiness of message traffic, network congestion often occurs. [0003] One way to solve network congestion is to limit the rate of the message at the receiving end. If the rate of the message is lower than the specified rate, the message will be received normally; if the rate of the message exceeds the specified rate, the message will be sent. The packet is discarded, or the priority of the packet is lowered. [0004] Currently, the commonly used method is to use token buckets to limit the rate of packets. As shown in Figure 1, the system continuously fills the token bucket with tokens at a specified rate until the token bucket is full. When a message arrives, compar...

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to provide a method of packet rate limiting based on token buckets, aiming at the inefficiency of the existing message rate limiting technology, which is not conducive to reducing the delay of message forwarding, etc., an improved message rate limit method is proposed. The rate limiting method divides the message into segments according to a fixed length, and limits the rate of the message in units of data segments; at the same time, a dual-port RAM (Random Access Memory: Random Access Memory) is used to count two consecutive data in the message The time difference between segments is used to calculate the number of filled tokens in the token bucket and reduce the forwarding delay time of messages in the system

Images