基于Go语言的点播云消息处理模块的设计与实现

发布时间：2018-06-27 11:11

  本文选题：点播云 + Go　； 参考：《南京大学》2017年硕士论文

【摘要】：随着互联网的大力发展,云服务逐渐普及起来。点播云是云计算在视频上的尝试,其背后蕴含了复杂的技术,包括上传下载,转码,域名分发,广告,加密防盗链等。点播云服务为中小型创业公司提供快速接入的开发工具,方便快速二次开发一个视频点播平台而不依赖于传统的大视频点播平台,能为用户提供便捷的视频服务。本文针对点播云业务场景,设计并实现了七牛云点播云平台的内部消息处理模块,以下简称Fetcher。本模块主要服务于点播云内部,针对用户上传视频文件后产生的消息数据进行处理操作。模块采用Go语言编写,同时使用了分布式消息队列Kafka和非关系型数据库MongoDB。本文主要阐述Fetcher模块及其高可用的设计与实现。本文首先介绍了点播云项目的背景和项目意义,描述了本系统使用的相关技术,包括Go语言技术栈,MongoDB,mgo框架,Heartbeat等。然后针对点播云的业务逻辑进行Fetcher的需求分析,对系统进行模块划分,介绍系统的整体架构,设计了数据库表,对各个模块的功能和业务流程进行了详细分析和设计,并设计和实现了 Fetcher模块的高可用。最后,本文用伪代码对模块进行描述,阐述了 Fetcher内部模块的具体实现过程,展示了部分模块的运行图。基于这些实现,能稳定支撑点播云的业务平台。目前有大量客户在使用七牛点播云。
[Abstract]:With the vigorous development of the Internet, cloud services are gradually popularizing. VOD cloud is an attempt of cloud computing in video, which contains complex technologies, such as uploading and downloading, transcoding, domain name distribution, advertising, encryption chain and so on. VOD cloud service provides a rapid access development tool for small and medium-sized startups. It is convenient to develop a video-on-demand platform quickly and twice without relying on the traditional VOD platform. It can provide users with convenient video services. In this paper, we design and implement the internal message processing module of QiNiuyun on demand cloud platform, which is referred to as Fetcher. This module is mainly used in the on-demand cloud to handle the message data generated after the user uploads the video file. The module is written in go language, using distributed message queue Kafka and non-relational database MongoDB. This paper mainly describes the design and implementation of Fetcher module and its high availability. This paper first introduces the background and significance of the on-demand cloud project, and describes the related technologies used in this system, including the go language technology stack, MongoDB / Mgo framework and Heartbeat, etc. Then the Fetcher requirement analysis is carried out for the business logic of the on-demand cloud, the system is divided into modules, the overall structure of the system is introduced, the database table is designed, and the functions and business processes of each module are analyzed and designed in detail. The high availability of Fetcher module is designed and implemented. Finally, this paper describes the module with pseudo code, describes the implementation process of Fetcher internal module, and shows the operation diagram of some modules. Based on these implementations, can support the cloud on demand business platform. At present, there are a large number of customers in the use of seven cattle on demand cloud.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.09
，

本文编号：2073729