Dynamic page loading method based on big data stream type calculation

Abstract

The invention discloses a dynamic page loading method based on big data stream type calculation. The dynamic page loading method comprises the following steps that (a) a key value pair relation table is established in a stream database, and the times of jumping from any one page to the other page on a website is stored; (b) the key value pair relation table of a current page clicked and opened by a user is searched in the stream database, and a group of pages with maximum relevance with the current page are selected out to serve as pre-loading pages; (c) a server reads the pre-loading pages and puts the pre-loading pages in a cache; (d) the server directly provides a generated page in the cache for the user if a predicted page is a next page needed to be jumped by the user. According to the dynamic page loading method, the stream type calculation database is adopted to record statistic page access relation data, the preloading pages are determined according to page relevance counts, judgment and dynamic preloading are performed according to statistical data of user behaviors, the dynamic page loading method is higher in applicability and accuracy, and the perception and experience of real users is greatly improved.

Description

Technical field [0001] The invention relates to a dynamic page loading method, in particular to a dynamic page loading method based on big data flow calculation. Background technique [0002] The current web server performance optimization technology is mainly divided into two types, the static page on the server side and the preloading on the client side. The page static technology mainly uses technical means to generate static pages from the data in the database, and regenerate the static pages when the data is refreshed, reducing the number of times to read the database to improve the performance of the web service. Client-side pre-loading technology uses the browser’s pre-loading function to predict the next page that the user will browse (or the next related page of a certain page can be specified by the developer), and the browser will browse the user The next page is pre-loaded on this page, so that when the user clicks on the next page, if the prediction is correct, the ...

AI Technical Summary

Problems solved by technology

[0003] The currently commonly used static loading process of pages is as follows: figure 1 As shown, it has the following defects: 1) a lot of files will be generated, and a large number of files will have an impact on storage space and performance; 2) page timeliness maintenance and management require a complex mechanism to achieve; 3) only use For pages that do not require high real-time performance, or have an obvious mechanism to trigger content modification (such as forums, news release pages, etc.), it is not suitable for content that may spontaneously change every minute and every second (such as monitoring data display pages, etc. )

[0004] The preloading technology based on browser prediction has the following defects: 1) whether the technology takes effect or not depends on whether the browser's prediction is accurate, and the browser's automatic prediction page is usually based on common page clues (such as next page, next category), which is not applicable to specific websites; 2) It is predicted by the website developer and assigned to the browser to preload, mainly relying on the developer's understanding of the page, because it is necessary to realize the location of the next page written in the code However, this method cannot dynamically adapt to changes in user browsing patterns, and manually predicting and writing codes for each different page will also increase the workload of development

Images