Hive Group By的实现原理

group by多字段

select rank, isonline, count(*) from city group by rank, isonline;

将GroupBy的字段组合为map的输出key值，利用MapReduce的排序，在reduce阶段保存LastKey区分不同的key。MapReduce的过程如下（当然这里只是说明Reduce端的非Hash聚合过程）

group by单字段

group by 单字段和多字段时的不同在于key上，以如下例子为例（出处太多）：
SELECT uid, SUM(COUNT) FROM logs GROUP BY uid;

hive> SELECT * FROM logs;
a   苹果  5
a   橙子  3
a      苹果   2
b   烧鸡  1

hive> SELECT uid, SUM(COUNT) FROM logs GROUP BY uid;
a   10
b   1

其过程如下图：

默认设置了hive.map.aggr=true，所以会在mapper端先group by一次，最后再把结果merge起来，为了减少reducer处理的数据量。注意看explain的mode是不一样的。mapper是hash，reducer是mergepartial。如果把hive.map.aggr=false，那将groupby放到reducer才做，他的mode是complete.

Explain

hive> explain SELECT uid, sum(count) FROM logs group by uid;
OK
ABSTRACT SYNTAX TREE:
  (TOK_QUERY (TOK_FROM (TOK_TABREF (TOK_TABNAME logs))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_TABLE_OR_COL uid)) (TOK_SELEXPR (TOK_FUNCTION sum (TOK_TABLE_OR_COL count)))) (TOK_GROUPBY (TOK_TABLE_OR_COL uid))))

STAGE DEPENDENCIES:
  Stage-1 is a root stage
  Stage-0 is a root stage

STAGE PLANS:
  Stage: Stage-1
    Map Reduce