CDN traffic anomaly detection device and method based on improved hierarchical temporal memory network

Abstract

The invention discloses a CDN flow anomaly detection device and method based on an improved layered time memory network. The device includes a data acquisition module, a data preprocessing module, a data storage module, a system scheduling module, an abnormality detection module and a display module. The method is: the data acquisition module collects the original log data, converts it into json format and sends it to the data preprocessing module; performs feature extraction to obtain the time series representation of CDN traffic, and the data storage module collects the log data of the data acquisition module and the data preprocessing module CDN data for storage; the anomaly detection module obtains traffic time series data through the system scheduling module, and inputs it into the time series anomaly detection model based on the improved hierarchical time memory network for online learning, completes the calculation of the anomaly possibility, and outputs the judgment of the anomaly possibility The detection results, the display module visualizes the key process. The invention has the advantages of fast detection speed and high accuracy.

Description

technical field [0001] The invention relates to the technical field of CDN traffic anomaly detection, in particular to a CDN traffic anomaly detection device and method based on an improved layered temporal memory network. Background technique [0002] In recent years, with the continuous improvement of Internet infrastructure, the digital strategy has been systematically explained, Internet services have continued to penetrate, and the number of Internet users has maintained a steady growth. In order to reduce the pressure on the network due to the rapid growth of user groups and huge data transmission volume, Content Delivery Network (CDN) came into being. CDN can serve the Internet in different locations through large-scale distributed deployment of server infrastructure. The inherent distribution of CDN brings popular applications and hot content as close as possible to users, which greatly reduces network delays, improves users' access speed and quality of experience, ...

AI Technical Summary

Problems solved by technology

[0004] 1. The execution efficiency needs to be further improved, and at the same time, it cannot meet the response time requirements for abnormal detection of large-traffic network links;

[0005] 2. In the detection algorithm, it is very dependent on the determination of the detection threshold, how to accurately calculate the threshold cannot give a good solution;

[0006] 3. The traditional method focuses on finding abnormalities, and has little involvement in diagnosing abnormal types

Images