基于Solr的混合介质存储在淘宝系统中的研究与应用  

Based on stored in Taobao Solr mixed media system in the research of application

作　　者：汪洋[1] 崔炜[2] 

机构地区：[1]烽火通信科技股份有限公司南京研发部,江苏南京210019 [2]武汉邮电科学研究院,湖北武汉430000

出　　处：《电子设计工程》2017年第15期22-25,共4页Electronic Design Engineering

基　　金：江苏省科技支撑计划(2015BAK20B01)

摘　　要：DT时代已经来临,开源搜索引擎已经成为各大公司检索海量数据的主流,其中Solr以其支持多种格式索引,高效性,高灵活性,高可扩展性,深受广大爱好者以及各大公司青睐,本文基于提高Solr检索速度和对数据管理便捷性,提出了混合介质存储分盘管理查询并且设置优先级异步并发的改进方案,通过在淘宝用户评论检索中的研究与应用,证实了方案能够极大优化了性能并且提高检索速度。DT times has come,open source search engine has become the mainstream of the major companies to retrieve massive data.Among them Solr to support a variety of formats index ,high efficiency,high flexibility,high saclability,bythe majority of fans and the favor of major companies.In this paper,base on the improved Solr retrieval speed and convenience data management ,put forward the mixed media storage,query points disk management and setting priorities asynchronous concurrent impovements.Through research and application in taobao users comments retrieval system ,confirmed the scheme to optimize performance and improve the retrieval speed.

关 键 词：LUCENE SOLR 混合介质存储存储 数据的冷热 淘宝用户评论检索系统 

分 类 号：TN02[电子电信—物理电子学]