HQL中count(*),count(1),count(column)的区别

原文地址：https://my.oschina.net/rathan/blog/608617

1、执行计划：

count(*)

explain select count(*) from dpods_SMS_Queue limit 100;
Explain
ABSTRACT SYNTAX TREE:
(TOK_QUERY (TOK_FROM (TOK_TABREF (TOK_TABNAME dpods_SMS_Queue))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_FUNCTIONSTAR count))) (TOK_LIMIT 100)))
STAGE DEPENDENCIES:
Stage-1 is a root stage
Stage-0 is a root stage
STAGE PLANS:
Stage: Stage-1
Map Reduce
Alias -> Map Operator Tree:
dpods_sms_queue
TableScan
alias: dpods_sms_queue
Select Operator
Group By Operator
aggregations:
expr: count() // 此处不同
bucketGroup: false
mode: hash
outputColumnNames: _col0
Reduce Output Operator
sort order:
tag: -1
value expressions:
expr: _col0
type: bigint
Reduce Operator Tree:
Group By Operator
aggregations:
expr: count(VALUE._col0)
bucketGroup: false
mode: mergepartial
outputColumnNames: _col0
Select Operator
expressions:
expr: _col0
type: bigint
outputColumnNames: _col0
Limit
File Output Operator
compressed: false
GlobalTableId: 0
table:
input format: org.apache.hadoop.mapred.TextInputFormat
output format: org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
Stage: Stage-0
Fetch Operator
limit: 100

count(1)

explain select count(1) from dpods_SMS_Queue limit 100;
Explain
ABSTRACT SYNTAX TREE:
(TOK_QUERY (TOK_FROM (TOK_TABREF (TOK_TABNAME dpods_SMS_Queue))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_FUNCTION count 1))) (TOK_LIMIT 100)))
STAGE DEPENDENCIES:
Stage-1 is a root stage
Stage-0 is a root stage
STAGE PLANS:
Stage: Stage-1
Map Reduce
Alias -> Map Operator Tree:
dpods_sms_queue
TableScan
alias: dpods_sms_queue
Select Operator
Group By Operator
aggregations:
expr: count(1) // 此处不同
bucketGroup: false
mode: hash
outputColumnNames: _col0
Reduce Output Operator
sort order:
tag: -1
value expressions:
expr: _col0
type: bigint
Reduce Operator Tree:
Group By Operator
aggregations:
expr: count(VALUE._col0)
bucketGroup: false
mode: mergepartial
outputColumnNames: _col0
Select Operator
expressions:
expr: _col0
type: bigint
outputColumnNames: _col0
Limit
File Output Operator
compressed: false
GlobalTableId: 0
table:
input format: org.apache.hadoop.mapred.TextInputFormat
output format: org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
Stage: Stage-0
Fetch Operator
limit: 100

count(column)

explain select count(msgid) from dpods_SMS_Queue limit 100;
Explain
ABSTRACT SYNTAX TREE:
(TOK_QUERY (TOK_FROM (TOK_TABREF (TOK_TABNAME dpods_SMS_Queue))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_FUNCTION count (TOK_TABLE_OR_COL msgid)))) (TOK_LIMIT 100)))
STAGE DEPENDENCIES:
Stage-1 is a root stage
Stage-0 is a root stage
STAGE PLANS:
Stage: Stage-1
Map Reduce
Alias -> Map Operator Tree:
dpods_sms_queue
TableScan
alias: dpods_sms_queue
Select Operator // 此处多了一个Select Operator
expressions:
expr: msgid
type: int
outputColumnNames: msgid
Group By Operator
aggregations:
expr: count(msgid) // 此处不同
bucketGroup: false
mode: hash
outputColumnNames: _col0
Reduce Output Operator
sort order:
tag: -1
value expressions:
expr: _col0
type: bigint
Reduce Operator Tree:
Group By Operator
aggregations:
expr: count(VALUE._col0)
bucketGroup: false
mode: mergepartial
outputColumnNames: _col0
Select Operator
expressions:
expr: _col0
type: bigint
outputColumnNames: _col0
Limit
File Output Operator
compressed: false
GlobalTableId: 0
table:
input format: org.apache.hadoop.mapred.TextInputFormat
output format: org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
Stage: Stage-0
Fetch Operator
limit: 100

从上面三个执行计划来看，只有两处不同：

count(column)的时候多了一个Select operator；
Group By Operator 的 aggregations过程略有不同；

2、结果：

count(*)：所有行进行统计，包括NULL行

count(1)：所有行进行统计，包括NULL行

count(column)：对column中非Null进行统计

count(distinct column)：对column中非Null进行去重统计