CDN peak flow prediction method and device and storage medium

Abstract

The embodiment of the invention relates to the technical field of networks, and discloses a CDN peak flow prediction method and device, and a storage medium. The method comprises the steps: obtaininghistorical flow values of all provinces; predicting a peak flow value of each province in future time according to the historical flow value of each province; accumulating the peak flow values of eachprovince in the future time to obtain a first predicted value of national peak flow in the future time; obtaining a national historical flow value; predicting a second prediction value of the national peak flow in the future time according to the national historical flow value; and determining a national peak flow prediction value in the future time according to the first prediction value and thesecond prediction value. The prediction effective time is long, and the obtained national peak flow prediction result is accurate.

Description

technical field [0001] The embodiments of the present invention relate to the field of network technologies, and in particular to a CDN peak traffic prediction method, device and storage medium. Background technique [0002] Content Delivery Network (CDN, Content Delivery Network) reduces core network traffic and central server load by enabling users to access content nearby, thereby greatly improving user experience, and technically solving problems due to small network bandwidth and user access volume. Poor service quality due to large network size, uneven distribution of outlets, and single point of failure. [0003] Currently, CDN-based traffic forecasting is divided into two categories: real-time forecasting and offline forecasting. [0004] Most of the more mature technologies currently belong to real-time prediction, such as establishing a prediction model based on system load, cache file request scheduling and other correlations, and adjusting real-time scheduling a...

AI Technical Summary

Problems solved by technology

[0006] However, the inventors found that there are at least the following problems in the prior art: the effective time of the existing real-time forecasting model prediction is relatively short, and the offline forecasting model is the national historical The peak flow input prediction model predicts the peak flow of the whole country, but because the flow values ​​in different parts of the country are different, the forecast result using the flow of the whole country is not accurate

Images