Oracle主键与复合主键的性能分析，主键和复合主键，查询性能相同（索引高度相同，恰当的运用索引）。主键和复合主键，(update,in
总结：
1、主键和复合主键，查询性能相同（索引高度相同，恰当的运用索引）。
2、主键和复合主键，(update,insert)性能不同(因为复合主键会用更多的块来创建索引，所以update,insert性能低)
实验思路：
1、 建立实验表，及主键，联合2个主键，联合3个主键
2、 查看索引的结构
3、查看条件相同的，执行计划（来确定主键和复合主键的效率）
一、 建立实验表；test1为单主键为1个column，test2为联合主键2个columns，test3为联合主键3个columns
sql> create table test1(a number,b number,c number,primary key(a));
table created.
sql> create table test2(a number,b number,c number,primary key(a,b));
table created.
sql> create table test3(a number,b number,c number,primary key(a,b,c));
table created.
二、 查看索引的结构
1、先查看一下建立的表对应的索引
sql> select index_name,table_name from user_indexes;
index_name table_name
------------------------------ ------------------------------
sys_c005198 test1
sys_c005199 test2
sys_c005200 test3
2、写个储存过程来给实验表插入数据
begin
for i in 1..10000 loop
insert into test1 values(i,i+1,i+2);
commit;
end loop;
end;
test1
sql>analyze index sys_c005198 validate structure;
index analyzed.
sql> select height,blocks,br_blks,lf_blks,lf_rows,del_lf_rows from index_stats ;
height blocks br_blks lf_blks lf_rows del_lf_rows
---------- ---------- ---------- ---------- ---------- -----------
2 24 1 18 10000 0
test2
sql> analyze index sys_c005199 validate structure;
index analyzed.
sql> select height,blocks,br_blks,lf_blks,lf_rows,del_lf_rows from index_stats ;
height blocks br_blks lf_blks lf_rows del_lf_rows
---------- ---------- ---------- ---------- ---------- -----------
2 32 1 23 10000 0
test3
sql>analyze index sys_c005200 validate structure;
index analyzed.
sql> select height,blocks,br_blks,lf_blks,lf_rows,del_lf_rows from index_stats ;
height blocks br_blks lf_blks lf_rows del_lf_rows
---------- ---------- ---------- ---------- ---------- -----------
2 40 1 28 10000 0
总结：根据b-tree索引的结构特点。说明主键和联合主键，同样的数据联合主键需要更多的资源来维护索引。（联合主键索引因为用了更多的块，所以update,insert会比主键索引慢一些。至于查询下面研究）
三、 查看相同情况下，主键的效率。
1、 语句都让其走index unique scan，看看效率：
test1
sql> select a from test1 where a=5555;
a
----------
5555
execution plan
----------------------------------------------------------
plan hash value: 2716871853
--------------------------------------------------------------------------------
-
| id | operation | name | rows | bytes | cost (%cpu)| time
|
--------------------------------------------------------------------------------
-
| 0 | select statement | | 1 | 13 | 1 (0)| 00:00:01
|
|* 1 | index unique scan| sys_c005198 | 1 | 13 | 1 (0)| 00:00:01
|
--------------------------------------------------------------------------------
-
predicate information (identified by operation id):
---------------------------------------------------
1 - access("a"=5555)
statistics
----------------------------------------------------------
1 recursive calls
0 db block gets
2 consistent gets
0 physical reads
0 redo size
405 bytes sent via sql*net to client
385 bytes received via sql*net from client
2 sql*net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
test2
sql> select a,b from test2 where a=5555 and b=5556;
a b
---------- ----------
5555 5556
execution plan
----------------------------------------------------------
plan hash value: 3210951477
--------------------------------------------------------------------------------
-
| id | operation | name | rows | bytes | cost (%cpu)| time
|
--------------------------------------------------------------------------------
-
| 0 | select statement | | 1 | 26 | 1 (0)| 00:00:01
|
|* 1 | index unique scan| sys_c005199 | 1 | 26 | 1 (0)| 00:00:01
|
--------------------------------------------------------------------------------
-
predicate information (identified by operation id):
---------------------------------------------------
1 - access("a"=5555 and "b"=5556)
statistics
----------------------------------------------------------
1 recursive calls
0 db block gets
2 consistent gets
0 physical reads
0 redo size
460 bytes sent via sql*net to client
385 bytes received via sql*net from client
2 sql*net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
test3
sql> select a,b,c from test3 where a=5555 and b=5556 and c=5557;
a b c
---------- ---------- ----------
5555 5556 5557
execution plan
----------------------------------------------------------
plan hash value: 1852305570
--------------------------------------------------------------------------------
-
| id | operation | name | rows | bytes | cost (%cpu)| time
|
--------------------------------------------------------------------------------
-
| 0 | select statement | | 1 | 39 | 1 (0)| 00:00:01
|
|* 1 | index unique scan| sys_c005200 | 1 | 39 | 1 (0)| 00:00:01
|
--------------------------------------------------------------------------------
-
predicate information (identified by operation id):
---------------------------------------------------
1 - access("a"=5555 and "b"=5556 and "c"=5557)
statistics
----------------------------------------------------------
1 recursive calls
0 db block gets
2 consistent gets
0 physical reads
0 redo size
515 bytes sent via sql*net to client
385 bytes received via sql*net from client
2 sql*net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
分析：通过执行sql走index unique scan索引的情况，分析执行计划得到的结果是主键和联合主键性能相同：
（我们关注的：
1 recursive calls
0 db block gets
2 consistent gets
0 physical reads
0 redo size
消耗一致和cost消耗一致。）
总结：主键和联合主键，应用b-tree索引的情况下，如果我们的索引高度相同,且正确的应用索引。这样的情况下我们查询性能是相同的。
欢迎大家给与纠正错误，，共同提升！
