Research And Implementation Of Cost-optimized Multi-cloud Live Video Distribution Strategy

With the rapid development of high-speed Internet access and popularization of high-performance smart devices,live broadcast service has been widely used in many scenarios.To guarantee quality of live broadcast service,live streams are transcoded to adapt to the diversity of terminal devices,and CDN service provide low-latency distribution.The previous way of building independent data centers by live broadcast service providers has the disadvantages of high cost and unscalable resources.The cloud platform has the characteristics of fast scalability and pay on demand.According to the highly dynamics of viewers’requests,flexible resources are provided to complete transcoding and CDN distribution.However,due to complex and changeable dynamic environment of multi-cloud,live broadcast service fluctuations in quality of service.Most of the existing live broadcast cost optimization methods consider transcoding or CDN distribution,and do not jointly consider the impact of transcoding and CDN distribution on cost and quality of service,which leads to increase of cost and decline of quality of service.Therefore,it is difficult to provide multi-cloud live video distribution service with cost optimization and quality of service guarantee.The purpose of this paper is to study the cost-optimized live video distribution strategy in multi-cloud environment.Firstly,the impact of video source,transcoding and CDN distribution on quality of live video distribution service,and proposes a multi-factor dynamic monitoring and evaluation method of live video distribution service based on the change of cloud service load in multi-cloud environment.Video source information is processed by video classification method based on temporal and spatial features,and multi-factor real-time evaluation method based on non-reference model is used to deal with transcoding and CDN distribution indicators.It realizes real-time and comprehensive evaluation of live video distribution service quality in multi-cloud environment.Secondly,this paper proposes a collaborative decision method of cloud transcoding and CDN distribution,defines the cost model quality of service model and constraint in multi-cloud environment,and implements the cost optimization method based on multi-agent model to obtain the collaborative decision of transcoding and CDN distribution,so as to achieve the goal of reducing the operational cost of the system as much as possible on the basis of ensuring quality of live broadcast service.Finally,this paper designs and implements a multi-cloud live video distribution management system,which provides functional modules such as live stream management,cloud transcoding,CDN distribution,service monitoring and evaluation and decision optimization,and tests the system functions and methods.The experimental results show that the proposed method can evaluate the live video distribution service quality real-time and accurately in the multi-cloud environment and effectively reduce operational cost of the system.