Current limiting method and device

Abstract

The embodiment of the invention provides a flow limiting method and device, and the method comprises the steps: enabling a root node to judge whether the root node is abnormal or not according to a received request message, enabling all leaf nodes to share the total resources provided by the root node if the root node is normal, determining an excess leaf node if the root node is abnormal, and carrying out the flow limiting of the excess leaf node. In the process, the root node is monitored, when the root node is abnormal, the leaf node is determined, and the flow of the leaf node is limited,so that the situation that the service cannot be normally carried out or is blocked is avoided. Moreover, if the root node is not abnormal, each leaf node shares the total resource provided by the root node, so that the service corresponding to each leaf node is normally carried out.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of data processing, and in particular, to a current limiting method and device. Background technique [0002] At present, in many scenarios, it is necessary to limit the flow of data to avoid traffic conflicts between different services. Among them, traffic conflict means that one service preempts the data traffic of another service, so that the other service cannot be performed normally or freezes. [0003] In order to avoid traffic conflicts between different services, the total traffic is generally allocated through technologies such as traffic control (TC), and different services correspond to different traffics, thereby avoiding traffic conflicts. For example, the total traffic value of application A and application B is 10K, the traffic threshold of application A is 8K, and the traffic threshold of application B is 2K, and application A and application B contain multiple services...

AI Technical Summary

Problems solved by technology

[0004] However, after allocating traffic in the above-mentioned way, it is very likely that a certain application will be limited in use

Continuing with the above example, if the 2K traffic of application B is not enough, and application A only uses 6K traffic, at this time, although application A has surplus traffic, based on the pre-distributed traffic, application B cannot use the traffic of application A. The remaining traffic, which in turn causes the business of application B to preempt traffic from each other, causing some or all of the business of application B to fail or freeze

Images