Network node selection method and device, and storage medium

Abstract

The invention discloses a network node selection method and device and a storage medium. According to the invention, the method comprises the steps: taking the terminal user information obtained fromthe service request initiated by the terminal user as the dimension information for selecting the target node, and selecting the target node according to the terminal user information and the node network quality mapping table pre-constructed based on any user dimension; determining which network node is selected as a target network node according to the business index mode values recorded under each network node under the dimension information corresponding to the terminal user information, so as to select the network node according to different business indexes without additionally deployinga monitoring machine, and selecting the target network node according to the multi-dimensional division standard, so the selected target network node is more suitable for providing service for a terminal user.

Description

technical field [0001] The embodiment of the present invention relates to the technical field of the Internet, and in particular to a method, device and storage medium for selecting a network node. Background technique [0002] The Content Delivery Network (CDN) is an intelligent virtual network built on the basis of the existing network, relying on the edge servers (also called: network nodes or CDN nodes) deployed in various computer rooms, so that end users can Obtain the required content, thereby reducing network congestion and improving the response speed of end-user access. Therefore, how to provide end users with a suitable network node with better network quality for service to reduce network delay is the key to CDN technology. At present, the selection method of network nodes is either based on the principle of geographical proximity, that is, the request of an end user in a certain area is directly dispatched to the network node closest to the end user who initiat...

AI Technical Summary

Problems solved by technology

[0003] However, in practical applications, geographical location and network delay cannot be equated. Even in computer rooms that are close to the same province, there are differences in network quality, and for end users in border areas of provinces, there are often network nodes closer to neighboring provinces. In addition, the network quality of end users using different operator networks to the same nearest network node is also very different, especially in the case of cross-network access by users of small operators, the difference will be more obvious

[0004] However, in the network node selection method based on the node network monitoring and detection mechanism, the accuracy of monitoring and detection needs to be based on a large number of monitoring machines. For some remote areas or small operators, network resources are scarce and it is difficult to purchase monitoring machines. The problem of lack of regional monitoring data leads to unreasonable selection results, and the introduction of monitoring machines also increases the implementation cost virtually; secondly, the network between monitoring machines and network nodes belongs to the link from the computer room to the computer room, and connects with end users to The complex network environment of network nodes is quite different, so the detected network delay is not accurate enough; moreover, there is still a certain difference between the detection initiated by the monitoring machine and the real end user request, and it is impossible to completely simulate the real request. Behavior

[0005] That is to say, whether the selection of network nodes is based on the principle of geographical proximity or the selection of network nodes based on the node network monitoring and detection mechanism, there are obvious defects, and it cannot truly and accurately reflect the network nodes to end users. network conditions, it is impossible to select network nodes with better network quality for end users to use

Images