join
语句:select u.name,a.orderid from order a join user b on a.uid=b.uid
group by
语句:select rank,isonline,count(*) from city group by rank,isonline
count distinct
语句:select dealid,count(distinct uid) num from order group by dealid
总结:count distinct场景下,在Map阶段无法利用combine对输出结果消重,必须将id作为Key输出,在Reduce阶段再对来自于不同Map Task、相同Key的结果进行消重,计入最终统计值。在count distinct场景下只有一个reduce