Bandwidth filtering method and device based on deep learning, server and storage medium

Abstract

The embodiment of the invention relates to the technical field of communication, and discloses a bandwidth filtering method and device based on deep learning, a server and a storage medium. Accordingto the invention, the bandwidth data of the server is obtained in real time; and the obtained bandwidth data is input into a preset deep neural network model, the result output by the deep neural network model is the output bandwidth data after the input bandwidth data is filtered, and the preset deep neural network model is obtained by training historical bandwidth data and the output bandwidth data after the historical bandwidth data is filtered. On-line real-time filtering is carried out by using a preset deep neural network model, learning can be carried out according to real-time changesof on-line bandwidth data, a large amount of human input is avoided, and time and energy consumption is reduced, so that the problem of time delay caused by filtering is reduced. And on the other hand, the trained deep neural network model is deployed on line for real-time filtering, so that the filtering effect on the bandwidth data of the server is improved.

Description

technical field [0001] Embodiments of the present invention relate to the field of communication technologies, and in particular, to a bandwidth filtering method, device, server, and storage medium based on deep learning. Background technique [0002] Network service providers and consumers can transmit data. Bandwidth is used to identify the data transmission capability of signal transmission, that is, the amount of data passing through the link per unit time. The smaller the bandwidth changes over time, the more stable the data transmission, and the higher the quality of service provided by the service provider. In order to ensure high-quality services, it is necessary to monitor the bandwidth of the server, formulate relevant speed limit policies according to the real-time bandwidth of the server, and detect whether there is an abnormality in the bandwidth data. In the case of , filter processing is performed on the detected bandwidth data, and the filtered bandwidth dat...

AI Technical Summary

Problems solved by technology

[0003] The inventors have found that there are at least the following problems in the related art: since bandwidth fluctuations on different servers are different, a single filtering algorithm is selected to filter the bandwidth of different types of servers with the same parameters, and the filtering effect is poor; if different parameters or use A variety of filtering algorithms filter the bandwidth of different types of servers. It is necessary to artificially try different parameters or different filtering algorithms to determine the appropriate filtering method for the bandwidth data of each type of server. Due to the large number of online servers, this artificial The tried-and-true approach takes a lot of energy and time

Images