分布式计算在券商营销推广活动中的应用  

Application of Distributed Computing in Marketing Activities of Securities Companies

作　　者：俞枫[1] 张忍 郑爽 YU Feng;ZHANG Ren;ZHENG Shuang(Guotai Junan Securities Co.,Ltd., Shanghai 200120)

机构地区：[1]国泰君安证券股份有限公司,上海200120

出　　处：《微型电脑应用》2019年第4期132-135,共4页Microcomputer Applications

摘　　要：为了应对券商行业线上营销推广活动大量用户参与带来的系统压力,满足线上营销推广活动灵活配置事件处理规则和动态修改事件处理规则的需要,提出了基于Kafka分布式消息队列、MongoDB分片副本集、Storm流式分布式计算框架、Zookeeper分布式节点配置同步服务的分布式营销活动事件处理规则定义及事件处理框架,对该框架的高可用、易扩展等特性进行阐述并进行性能对比、验证。In order to relieve the system pressure caused by the large number of users participating in the online marketing activities of securities company and meet the need of flexibly configuring event processing rules and dynamically modifying event processing rules for the online marketing activities, this paper proposes a software framework for definition of distributed marketing activities’ rules and processing marketing activities’ events depending on Kafka, Mongo DB, Storm and Zookeeper. At the same time, we explain and verify the framework’s features such as the high availability and easy expansion.

关 键 词：STORM Kafka MongoDB分片副本集 Zookeeper 活动规则动态配置 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]