Bandwidth multiplexing method and device

Abstract

The invention discloses a bandwidth multiplexing method and device. The method includes the following steps: acquiring bandwidth time domain information of an obtained domain name, wherein the bandwidth time domain information includes bandwidth time domain information of each operator in each region; determining a bandwidth interval segment to which the bandwidth time domain information of different domain names corresponding to each operator in each region belongs; performing bandwidth combination for a target area, and combining the bandwidth time domain information of each domain name located in each bandwidth interval segment smaller than the bandwidth upper limit of the target area; and performing domain name access scheduling according to a bandwidth combination result. The scheme of the invention takes into account the factors of bandwidth peak values of different users, segments the bandwidth intervals, combines the bandwidth time domain information of each domain name of thebandwidth interval segment, and reasonably combines the time domain information; and the scheme also adopts bandwidth frequency domain information of the domain names, the data processing amount can be effectively reduced, the peak values determined according to the frequency domain information can be more accurate, and the accuracy of judgment of bandwidth peak periods can be improved.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a bandwidth multiplexing method and device. Background technique [0002] In the existing cloud distribution environment, for example, in the content delivery network (Content Delivery Network, referred to as CDN), the bandwidth consumed by users when using network services is usually billed by a fixed duration, such as monthly billing or billed annually. In practical applications, the cloud distribution environment serves thousands of customers, and each customer uses different services, so the bandwidth model is also different. Therefore, how to select the optimal combination scheme from the bandwidth model of each customer, improve bandwidth reuse rate, increase profit margin and save cost is an urgent problem to be solved. [0003] In the prior art, by using the initial bandwidth model and log data analysis, combined with the relationship between each service, the law ...

AI Technical Summary

Problems solved by technology

[0005] (1) In this method, time-domain signals are generally used. The time-domain signals include signal values ​​recorded every 1 minute or 5 minutes. When there are many network nodes, the calculation amount is huge when directly using the time-domain signals for calculation, and the processing effect is low.

[0006] (2) When scheduling according to the bandwidth of different peak periods, only the billing characteristics and bandwidth peak data of each cloud distribution node server corresponding to the network resource data in the CDN network are obtained, and the data related to bandwidth configuration is generated through these data, and then Carry out corresponding scheduling. These data combinations themselves are not real-time dynamic, but are calculated and evaluated over a period of time. These data combinations will affect the accuracy of scheduling

[0007] (3) After the verification of practical application, the combination effect of the final combination method obtained by this method is not ideal

Images