CN-122027490-A - Bandwidth cost analysis method, device, equipment and medium for content distribution network

CN122027490ACN 122027490 ACN122027490 ACN 122027490ACN-122027490-A

Abstract

The application provides a bandwidth cost analysis method, device, equipment and medium of a content distribution network, and relates to the technical field of network resource cost management. The method comprises the steps of obtaining service labels of a plurality of historical request logs in each service dimension, conducting time aggregation on the plurality of historical request logs to obtain request log sets corresponding to a plurality of historical time periods, conducting aggregation on the historical request logs in the request log sets corresponding to each historical time period according to each target service label under a target service dimension combination to obtain bandwidth aggregation data of the target service dimension combination in each historical time period, and determining bandwidth cost of a content distribution network under the target service dimension combination according to the bandwidth aggregation data of the target service dimension combination in each historical time period and cost information of each historical time period. The application can aggregate bandwidth aggregation data and cost information of the target service dimension combination, and provides data support for cost analysis.

Inventors

LI WENYAN
CHEN SONGMAO
XIE HAOJIE
WEN PAN
WANG WEIHENG
ZENG SHAOBIN
CHEN JINHAI

Assignees

杭州网易云音乐科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (10)

1. A method for analyzing bandwidth costs of a content distribution network, comprising: Acquiring a plurality of historical request logs of a content distribution network; Analyzing a plurality of service dimensions of the history request logs to obtain service labels of the history request logs in each service dimension; Performing time aggregation on a plurality of historical request logs to obtain request log sets corresponding to a plurality of historical time periods; According to each target service label under the target service dimension combination, aggregating the history request logs in the request log set corresponding to each history time period to obtain bandwidth aggregation data of the target service dimension combination in each history time period; acquiring cost information of the content distribution network in each historical time period; And determining the bandwidth cost of the content distribution network under the target service dimension combination according to the bandwidth aggregation data of the target service dimension combination in each historical time period and the cost information of each historical time period.
2. The method of claim 1, wherein the plurality of business dimensions comprises a business line dimension and a business scenario dimension; The analyzing the plurality of history request logs in a plurality of service dimensions to obtain service labels of the plurality of history request logs in each service dimension comprises the following steps: And determining service line labels of the history request logs in the service line dimension and service scene labels of the history request logs in the service scene dimension according to the request resource positioning information in the history request logs.
3. The method of claim 1, wherein the plurality of service dimensions further comprises a client version dimension and a terminal type dimension; The analyzing the plurality of history request logs in a plurality of service dimensions to obtain service labels of the plurality of history request logs in each service dimension comprises the following steps: And determining client version labels of the history request logs in the client version dimension and terminal type labels in the terminal type dimension according to the user agent information of the history request logs.
4. The method of claim 1, wherein the plurality of business dimensions further comprises an experimental dimension and a custom dimension; The analyzing the plurality of history request logs in a plurality of service dimensions to obtain service labels of the plurality of history request logs in each service dimension comprises the following steps: And determining experimental labels of the history request logs in the experimental dimension and business labels in the custom dimension according to the custom request fields of the history request logs.
5. The method of claim 1, wherein if the plurality of service dimensions includes a terminal type dimension and an experiment dimension, the method further comprises: acquiring a large disc bandwidth of the content distribution network in the historical time period, wherein the large disc bandwidth is the sum of preallocated bandwidths of all terminal types in the historical time period; Acquiring pre-allocation bandwidths of target terminal types in all terminal types; according to the experiment label in the experiment dimension and the label of the target terminal type in the terminal type dimension, analyzing the request log set corresponding to the historical time period to obtain the bandwidth of each experiment group under the target terminal type; And calculating the large disc bandwidth of the experiment group in the historical time period according to the large disc bandwidth of the historical time period, the bandwidth of the experiment group in the target terminal type, the pre-allocation bandwidth of the target terminal type and the target push full proportion.
6. The method according to claim 5, wherein the analyzing the request log set corresponding to the historical time period according to the experiment label in the experiment dimension and the label of the target terminal type in the terminal type dimension to obtain the bandwidth of the experiment group in the target terminal type includes: Grouping request log sets corresponding to the historical time periods according to experimental labels in experimental dimensions to obtain request logs of a plurality of experimental groups; and acquiring the bandwidth of the experimental group under the target terminal type according to the label of the target terminal type under the terminal type dimension and the request log of the experimental group.
7. The method of claim 5, wherein the method further comprises: determining a large disc bandwidth peak value of the experimental group according to the large disc bandwidths of the experimental group in a plurality of historical time periods in a preset observation period; and determining a target experimental group according to the large disc bandwidth peaks of a plurality of experimental groups.
8. A bandwidth cost analysis apparatus of a content distribution network, comprising: A first acquisition module for acquiring a plurality of history request logs of the content distribution network; the analysis module is used for analyzing a plurality of service dimensions of the history request log to obtain service labels of the history request log in each service dimension; the first aggregation module is used for carrying out time aggregation on a plurality of historical request logs to obtain a request log set corresponding to a plurality of historical time periods; the second aggregation module is used for aggregating the historical request logs in the request log set corresponding to each historical time period according to each target service label under the target service dimension combination to obtain bandwidth aggregation data of the target service dimension combination in each historical time period; a second obtaining module, configured to obtain cost information of the content distribution network in each of the historical time periods; And the determining module is used for determining the bandwidth cost of the content distribution network under the target service dimension combination according to the bandwidth aggregation data of the target service labels in the historical time periods and the cost information of each historical time period.
9. An electronic device comprising a processor, a storage medium and a bus, the storage medium storing program instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is operating, the processor executing the program instructions to implement the bandwidth cost analysis method of a content delivery network according to any one of claims 1 to 7.
10. A readable storage medium having stored thereon program instructions which when executed by a processor implement the bandwidth cost analysis method of a content distribution network according to any of claims 1 to 7.

Description

Bandwidth cost analysis method, device, equipment and medium for content distribution network Technical Field The present application relates to the field of network resource cost management technologies, and in particular, to a method, an apparatus, a device, and a medium for analyzing bandwidth cost of a content distribution network. Background With the rapid development of internet technology, a content delivery network (Content Delivery Network, CDN) is used as a core infrastructure for improving network resource transmission efficiency and optimizing user access experience, and has been widely applied to various service scenarios such as video playing, file downloading, web acceleration, and the like. Meanwhile, the CDN bandwidth cost also becomes one of the core operation costs of the Internet enterprises, and the CDN bandwidth cost is accurately analyzed and efficiently managed and controlled, so that the operation benefit and the market competitiveness of the enterprises are directly related. The prior CDN bandwidth cost is a coarse particle cost statistics scheme according to domain name/bill dimension, and the scheme mainly depends on bills and simple flow statistics reports provided by CDN service providers, and statistics of macroscopic indexes such as total cost, domain name cost ratio and the like in a certain account period is realized by importing bill data into an internal financial system. When the specific service cost needs to be analyzed, the domain name and the service line are mapped roughly in a manual mode, and temporary cost aggregation is completed. However, since a single domain name usually bears a plurality of service lines or service scenes, the method cannot distinguish cost contributions of different services under the same domain name, so that cost amortization deviation of the service lines is larger, and the method lacks distinguishing capability of experimental dimensions, cannot support disassembling cost according to experimental grouping, and cannot evaluate cost differences after the whole amount of different experimental schemes are pushed away. In summary, the prior art solution stays in the coarse-grain cost statistics layer, and cannot meet the requirement of enterprises for fine cost attribution in CDN bandwidth cost management, so a technical solution for quantitatively analyzing CDN bandwidth cost is needed. Disclosure of Invention The present application addresses the above-mentioned shortcomings in the prior art by providing a method, apparatus, device, and medium for analyzing bandwidth cost of a content distribution network, so as to solve the problems existing in the prior art. The technical scheme adopted by the embodiment of the application is as follows: in a first aspect, an embodiment of the present application provides a bandwidth cost analysis method of a content distribution network, including: Acquiring a plurality of historical request logs of a content distribution network; Analyzing a plurality of service dimensions of the history request logs to obtain service labels of the history request logs in each service dimension; Performing time aggregation on a plurality of historical request logs to obtain request log sets corresponding to a plurality of historical time periods; According to each target service label under the target service dimension combination, aggregating the history request logs in the request log set corresponding to each history time period to obtain bandwidth aggregation data of the target service dimension combination in each history time period; acquiring cost information of the content distribution network in each historical time period; And determining the bandwidth cost of the content distribution network under the target service dimension combination according to the bandwidth aggregation data of the target service dimension combination in each historical time period and the cost information of each historical time period. In one embodiment, the plurality of business dimensions includes a business line dimension and a business scenario dimension; The analyzing the plurality of history request logs in a plurality of service dimensions to obtain service labels of the plurality of history request logs in each service dimension comprises the following steps: And determining service line labels of the history request logs in the service line dimension and service scene labels of the history request logs in the service scene dimension according to the request resource positioning information in the history request logs. In one embodiment, the plurality of service dimensions further includes a client version dimension and a terminal type dimension; The analyzing the plurality of history request logs in a plurality of service dimensions to obtain service labels of the plurality of history request logs in each service dimension comprises the following steps: And determining client version labels of the history request logs in t