2、公安局道路车辆监控数据三年可达200亿条、总量120TB。据世界权威IT信息咨询分析公司IDC研究报告预测：全世界数据量未来10年将从2009年的0.8ZB增长到2020年的35ZB(1ZB=1000EB=1000000PB)，10年将增长44倍，年均增长40%。由于数据量的快速增长，对大数据的操作和结构化查询在日常的数据处理经常用到，聚集查询也是查询时使用比较多的查询。关键词：聚集查询;结构化查询IVABSTRACTIn recent years, with the rapid developmentof computerand informationte
3、chnology，, industry application system has expanded , andthedatageneratedbytheapplciationsgrowsfast. Thedatawhichalways reaches hundreds of TB or tens to hundreds of PBhas been far beyond the existing traditional processing capacity of information system. Therefore, to seek effect
4、ive data processing technology, method and means isingreatneedin the real world.Baidu now ownsdataexceeded100PB,andithasto deal with data withvolumeof 10 PB ~ 100 PB; thetransactiondataofTaobaoreaches 100 PB; Twitter releases more than 200 million messages a day; Sina Weibo posts
5、 80 millionmessageaday; datageneratedofone province of China Mobile Communications telephoneis up to 0.5PB ~ 1PBamonth; a capital city theroad vehicle monitoring data ofacapitalcityinthreeyearsis up to 120TB. According to the world authority IT information consulting research firm
6、 IDC analysis report, the amount of data generatedbythewholeworldin the next 10 years will increase from 0.8ZBin2009to 35ZB 35 in 2020(1ZB = 1000EB = 1000000PB).Thedatavolume grows 44 timesintenyears, with an average annual growth of 40%. Dueto the rapid growth in data volume, t
7、he operation of the large data structured query is often used in the daily data processing, aggregate query is oneofthequeriesusedmostlyinbigdataprocessing.Keywords：AggregateQuery;StructuredQueryIV1绪论11.1本文研究的背景和意义11.2国内外研究现状与热点41.2.1大数据研究文献的国别和机构分布41.2.2大数据研究的学科领域分布51.2.3大数据产业技术创新
