36.读书笔记收获不止Oracle之表哈希、合并排序连接与索引

前端之家收集整理的这篇文章主要介绍了36.读书笔记收获不止Oracle之表哈希、合并排序连接与索引前端之家小编觉得挺不错的,现在分享给大家,也给大家做个参考。

36.读书笔记收获不止Oracle之表哈希、合并排序连接与索引

1. 哈希连接与索引

对于哈希连接和排序合并连接,索引的连接条件起不到快速检索的作用。但是限制条件列如果有合适的索引可以快速检索到少量记录,是可以提升性能的。

在没有任何索引的情况下,ORACLE倾向走哈希连接,因为哈希连接算法本身还是比较先进的。哈希连接需要在PGA中的HASH_AREA_SIZE中完成。

2. 合并排序连接与索引

索引对哈希连接来说,仅仅是考虑限制条件上的索引是否能用上索引,连接条件上的索引是不能发挥作用的。排序合并连接和哈希连接有差别,排序合并连接上的连接条件没有检索作用,却有消除排序的作用。

实验如下:

sql>alter session set statistics_level=all;

sql>select /*+ordered use_merge(t2)*/ * from t1,t2 wheret1.id=t2.t1_id;

sql> select * fromtable(dbms_xplan.display_cursor(null,null,'allstats last'));

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

sql_ID 5jxwkr07a1jdp,child number 0

-------------------------------------

select /*+ordered use_merge(t2)*/ * fromt1,t2 where t1.id=t2.t1_id

Plan hash value: 412793182

--------------------------------------------------------------------------------

---------------------------------

| Id| Operation | Name | Starts | E-Rows | A-Rows | A-Time| Buf

fers | OMem| 1Mem | Used-Mem |

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

---------------------------------

| 0| SELECT STATEMENT | | 1 | | 100 |00:00:00.05 |

1012 | | | |

| 1| MERGE JOIN | | 1 | 100 | 100|00:00:00.05 |

1012 | | | |

| 2| SORT JOIN | | 1 | 100 | 100|00:00:00.01 |

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

7| 13312 | 13312 |12288 (0)|

| 3| TABLE ACCESS FULL| T1 | 1 | 100 | 100 |00:00:00.01 |

7| | | |

|* 4| SORT JOIN | | 100 | 100K| 100 |00:00:00.05 |

1005 | 9762K| 1209K| 8677K (0)|

| 5| TABLE ACCESS FULL| T2 | 1 | 100K| 100K|00:00:00.01 |

1005 | | | |

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

---------------------------------

Predicate Information (identified byoperation id):

---------------------------------------------------

4- access("T1"."ID"="T2"."T1_ID")

filter("T1"."ID"="T2"."T1_ID")

23 rows selected.

建索引观察,在T1表的连接条件ID列上建索引,然后观察。

Create index idx_t1_id on t1(id);

sql>select /*+ordered use_merge(t2)*/ * from t1,t2 wheret1.id=t2.t1_id;

sql> select * from table(dbms_xplan.display_cursor(null,'allstatslast'));

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

sql_ID 5jxwkr07a1jdp,t2 where t1.id=t2.t1_id

Plan hash value: 2678642687

--------------------------------------------------------------------------------

-----------------------------------------------

| Id| Operation | Name | Starts | E-Rows | A-Rows |

A-Time | Buffers | OMem | 1Mem | Used-Mem |

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

-----------------------------------------------

| 0| SELECT STATEMENT | | 1 | | 100 |00:

00:00.06 | 1021 | | | |

| 1| MERGE JOIN | | 1 | 100 | 100|00:

00:00.06 | 1021 | | | |

| 2| TABLE ACCESS BY INDEX ROWID| T1 | 1| 100 | 100 |00:

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

00:00.01 | 16| | | |

| 3| INDEX FULL SCAN | IDX_T1_ID | 1 | 100 | 100|00:

00:00.01 | 8 | | | |

|* 4| SORT JOIN | | 100 |100K| 100 |00:

00:00.06 | 1005 |9762K| 1209K| 8677K (0)|

| 5| TABLE ACCESS FULL | T2 | 1 | 100K| 100K|00:

00:00.02 | 1005 | | | |

PLAN_TABLE_OUTPUT

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

-----------------------------------------------

Predicate Information (identified byoperation id):

---------------------------------------------------

4- access("T1"."ID"="T2"."T1_ID")

filter("T1"."ID"="T2"."T1_ID")

23 rows selected.

对连接条件列建索引,可以消除一张表的排序,提升效率。

此外还可以增大内存排序区,避免在排序尺寸过大时在磁盘中排序。

原文链接:https://www.f2er.com/oracle/207042.html

猜你在找的Oracle相关文章