A multi-data center-based optimization method for uploading code rate of cloud live streaming

Abstract

The present invention relates to a multi-data center-based cloud live broadcast upload code rate optimization method, comprising the following steps: S1. Obtain system information, including the number of data centers and the maximum bandwidth limit, the minimum and maximum bandwidth limit of the uploader, the viewing user and The number of uploaders, set the stop condition; S2. Solve the minimum and maximum total upload bandwidth of the cloud live broadcast platform, and determine the value of the upload bandwidth of each data center; S3. Determine the data of the uploader according to the value of the upload bandwidth of each data center Center selection and upload bit rate selection. At this time, the uploader may choose more than one data center; S4. According to the optimal decision determined in the previous step, adjust the selection of each uploader so that each uploader can only choose one Data center; S5. Judging whether the stop condition has been reached, if yes, skip to step S6, otherwise skip to step S2; S6. Output the global optimal total upload bandwidth, each uploader's data center selection and upload code rate.

Description

technical field [0001] The present invention relates to the fields of multimedia network and cloud computing resource management, and more specifically, to a multi-data center-based optimization method for cloud live upload code rate. Background technique [0002] In recent years, with the rise of cloud live broadcast services and the continuous upgrading of client devices, users can broadcast their own recorded videos to others through the cloud live broadcast platform anytime and anywhere. Many cloud live broadcast platforms have emerged in China, such as Panda Live, Comprehensive Live, and Huya Live, which occupy a large amount of video network traffic. Taking Panda Live as an example, its content involves popular game live broadcasts, talent live broadcasts, and sports live broadcasts. In the cloud live broadcast service, it mainly includes three roles: the cloud live broadcast platform composed of multiple data centers, the uploader, and the viewer. Its general struct...

AI Technical Summary

Problems solved by technology

[0003] However, there are still many challenges in the current cloud live broadcast service architecture

With the rise of cloud live broadcast services, researchers in the industry and academia are gradually focusing on the optimization and performance improvement of cloud live broadcast architecture, but there are the following problems: 1) Most of the existing research focuses on data centers and Optimization on the download side, such as data center transcoding and traffic migration, video stream distribution and content distribution network acceleration on the download side, but few studies focus on the optimization of the upload code rate of the cloud live broadcast service; 2) when it comes to cloud live broadcast In the optimization work of the upload end of the service, the existing research either roughly divides the uploader into different regions, or restricts the user to a single data center, which has great limitations; 3) In the existing cloud live broadcast In service uploading, the total upload bandwidth is set based on experience, lacking theoretical guidance, which can easily lead to redundancy and waste of bandwidth

Images