CDN prefix-based cache clearing method and system

Abstract

The invention relates to a CDN prefix-based cache clearing system. The system comprises a data preprocessing module which has the functions of domain name prefix list extraction, refreshing and the like, a storage module which has a domain name list storage function, prefix refreshing task time and a caching time storage function, and a data analyzing processing module used for performing analyzing processing on file prefix data and obtaining a file from a source webpage once again. According to the CDN cache clearing method, a domain name maintaining prefix refreshing list is stored to each machine, and the time of a prefix refreshing task is recorded; for each file, the time when a file is cached is stored in meta data of the file; every time when a request arrives, whether or not a prefix refreshing list is hit is compared, if the prefix refreshing list is hit, the caching time of a file is queried, and if the refreshing time is latter than the caching time, the file is cleared anda previous level is returned to obtain the file once again; a full deletion mechanism is replaced with a marking function, and an action in which clearing is conducted according to prefixes can becomeeffective rapidly.

Description

technical field [0001] The invention belongs to the field of information technology, and in particular relates to a method and system for clearing CDN buffers by prefixes. Background technique [0002] In distributed systems such as CDN, content is stored in CDN nodes through a cache mechanism, and users obtain data through CDN nodes. Because the CDN node is cached, there may be data inconsistencies between the CDN node and the origin site. In actual business, it is often encountered the function of clearing the specified file cache in the CDN, so as to achieve the purpose of complete data synchronization between the CDN and the origin site. In actual use, there are certain rules in the directory structure of the website, and the function of updating all files under a specified prefix is ​​often encountered (such as: clearing all files under a specific directory). In the mechanism of clearing by prefix, CDN nodes are required to be able to store the corresponding relations...

AI Technical Summary

Problems solved by technology

Disadvantages of the traditional solution: 1. Record the access log for each request, and sort the access log by prefix, which has high performance overhead; 2. Every time a prefix is ​​cleared, all files corresponding to the prefix need to be cleared immediately, and the action of clearing consumes a lot , and the clearing takes a long time to take effect

Images