两张表合并结果，求一个性能较好的sql

[liquidcn]

表A:

C1 C2 C3 C4
a b c 10
b c d 5
e a f 12
a b c 11
e a f 8
f e b 33
b e h 50

表B：

C1 C2 C3 C5
e a f 10
b c d 5
e a f 12
a b c 11
b c d 8


合并成表C:

C1 C2 C3 C4 C5
a b c 21 11
b c d 5 13
e a f 20 21
f e b 33
b e h 50


能想到的就是先group表A到临时表T，然后遍历表C，根据条件将B group后的结果插入C。。。不过貌似太low了有点，性能也差
所以求教一个性能好的sql

[liquidcn]

版主就是不一样，各种效率！！

[liquidcn]

临时表C，不是T

[發糞塗牆]

union all 再在外层sum

[發糞塗牆]

----------------------------------------------------------------
-- Author  :DBA_HuangZJ(發糞塗牆)
-- Date    :2014-09-29 12:23:46
-- Version:
--      Microsoft SQL Server 2012 - 11.0.5058.0 (X64) 
--	May 14 2014 18:34:29 
--	Copyright (c) Microsoft Corporation
--	Enterprise Edition: Core-based Licensing (64-bit) on Windows NT 6.3 <X64> (Build 9600: ) (Hypervisor)
--
----------------------------------------------------------------
--> 测试数据：[A]
if object_id('[A]') is not null drop table [A]
go 
create table [A]([C1] varchar(1),[C2] varchar(1),[C3] varchar(1),[C4] int)
insert [A]
select 'a','b','c',10 union all
select 'b','c','d',5 union all
select 'e','a','f',12 union all
select 'a','b','c',11 union all
select 'e','a','f',8 union all
select 'f','e','b',33 union all
select 'b','e','h',50
--> 测试数据：[B]
if object_id('[B]') is not null drop table [B]
go 
create table [B]([C1] varchar(1),[C2] varchar(1),[C3] varchar(1),[C5] int)
insert [B]
select 'e','a','f',10 union all
select 'b','c','d',5 union all
select 'e','a','f',12 union all
select 'a','b','c',11 union all
select 'b','c','d',8
--------------开始查询--------------------------


SELECT c1,c2,c3,SUM(c4)c4,SUM(c5)c5
FROM (
select C1  ,   C2  ,    C3,        C4    , 0 AS   C5 from [A]
UNION ALL 
select C1  ,   C2  ,    C3,   0 AS      C4    ,  C5 from [B])a
GROUP BY c1,c2,c3
----------------结果----------------------------
/* 
c1   c2   c3   c4          c5
---- ---- ---- ----------- -----------
a    b    c    21          11
b    c    d    5           13
b    e    h    50          0
e    a    f    20          22
f    e    b    33          0
*/