Method and device for dividing filling rate interval based on token bucket algorithm

Abstract

The embodiment of the invention provides a method and a device for dividing a filling rate interval based on token bucket algorithm. The method comprises the following steps: acquiring a corresponding decimal data bit width M when the number of tokens in a token bucket is minimum and a corresponding integral data bit width N when the number of the tokens is maximum in the filling rate interval, wherein the M and the N are positive integers; selecting a larger data bit width in the data bit width M and the data bit width N; and dividing the filling rate interval into a plurality of subintervals according to the larger data bit width, wherein the data bit width corresponding to the number of the tokens in the token bucket in the subintervals is the larger data bit width. By dividing the filling rate interval into the plurality of subintervals, the packet processing time interval for network equipment is smaller; and when the filling rate range is larger, the requirement of flow control can be met by a Memory with smaller data bit width; therefore, the embodiment of the invention reduces circuit resources for realizing the flow control so as to greatly reduce the cost.

Description

technical field [0001] The invention relates to the field of data communication, in particular to a method and device for dividing a filling rate interval based on a token bucket algorithm. Background technique [0002] In data communication technology, flow control is a key technology to realize Quality of Service (QOS), through flow control, users can be provided with different access rates and burst bandwidths. At present, the flow control in the data communication system is basically realized based on the RFC2697 / RFC2698 token bucket algorithm. [0003] Taking the tcTCM token bucket algorithm specified in RFC2698 as an example, the parameters to implement the algorithm include: Committed Information Rate (CIR), Committed Burst Size (CBS), Peak Information Rate (Peak Information Rate, PIR) and Peak Burst Size (Peak Burst Size, PBS), where CBS represents the maximum depth of the C bucket, PBS represents the maximum depth of the P bucket, CIR is the speed of adding tokens ...

AI Technical Summary

Problems solved by technology

[0008] Since both Tc(tn) and Tp(tn) are stored in external flash memory (Memory), Tc(tn) and Tp(tn) with a large data bit width require large-capacity Memory to support them; in addition, for traffic In the circuit for specific implementation of the control, since it involves many arithmetic operation logics, Tc(tn) and Tp(tn) with a large data bit width will consume more circuit resources

In general, Tc(tn) and Tp(tn) with a large data bit width lead to more costly implementation of flow control

Images