【MOS】Limitations of the Oracle Cost Based Optimizer (文档 ID 212809.1)

简介: 【MOS】Limitations of the Oracle Cost Based Optimizer (文档 ID 212809.1) APPLIES TO: Oracle Database - Personal Edition - Version 7.

【MOS】Limitations of the Oracle Cost Based Optimizer (文档 ID 212809.1)




APPLIES TO:

Oracle Database - Personal Edition - Version 7.1.4.0 and later
Oracle Database - Enterprise Edition - Version 6.0.0.0 and later
Oracle Database - Standard Edition - Version 7.0.16.0 and later
Information in this document applies to any platform.

PURPOSE

This Document outlines some of the limitations that may be encountered when using the Cost Based Optimizer.

SCOPE

Cost Based Optimizer Limitations

The Cost Based Optimizer (CBO) uses a complex and comprehensive model to choose the plan with the lowest cost overall. In most cases the model picks the best access methods for accessing the data in the most efficient manner. However, even with computed statistics and column histograms it is possible for the Cost Based Optimizer to choose a sub-optimal plan. There are limitations to the Cost model that can affect queries in some circumstances. Some of these are listed and explained below:

DETAILS

  • Potential for incorrect estimation of intermediate result set cardinality

    Cardinality is the CBO estimate of the number of rows produced by a row source or combination of row sources. In some cases, the cardinality of result sets can be miscalculated. This is most common with complex predicates where the statistics do not accurately reflect the data or where predicate values are correlated. The following is an illustration of a statement featuring correlated predicates:

    >

         >   

     

      

     
  • <><> 



  •   

  •  

     

     

     

     

     

     
     
     

     

  •  

  • <> 

     

     


 
 
 
 

 
 
 
 

 







 

    

img_e3029f287d989cd04bd75432ecc1c172.png
DBA笔试面试讲解
欢迎与我联系

目录
相关文章
|
SQL 分布式计算 运维
【大数据开发运维解决方案】Sqoop增量同步mysql/oracle数据到hive(merge-key/append)测试文档
上一篇文章介绍了sqoop全量同步数据到hive, 本片文章将通过实验详细介绍如何增量同步数据到hive,以及sqoop job与crontab定时结合无密码登录的增量同步实现方法。
【大数据开发运维解决方案】Sqoop增量同步mysql/oracle数据到hive(merge-key/append)测试文档
|
SQL Oracle 关系型数据库
使用PL/SQL Developer给Oracle生成漂亮的数据库说明文档
使用PL/SQL Developer给Oracle生成漂亮的数据库说明文档
228 0
使用PL/SQL Developer给Oracle生成漂亮的数据库说明文档
|
缓存 监控 Oracle
Oracle检查点(Checkpoint)详解 (文档 ID 1526118.1)
Oracle检查点(Checkpoint)详解 1.检查点概念--chkpoint 检查点是一个数据库事件,存在的意义在于减少崩溃恢复crash recovery时间.检查点事件由后台进程CKPT触发,当检查点发生时,CKPT通知DBWR进程将脏数据库dirtybuffer写出到数据文件上,更新数据文件头及控制文件上的检查点信息。
2689 0
|
Oracle 关系型数据库 MySQL
根据阿里文档定义oracle建表规范
根据阿里文档定义oracle建表规范
310 0
根据阿里文档定义oracle建表规范
|
Oracle 关系型数据库 数据库
|
存储 Oracle 关系型数据库
Oracle 11gR2 restart 单机使用asm存储 主机名发生更改处理过程 (文档 ID 986740.1)
How to Reconfigure Oracle Restart (文档 ID 986740.1) In this Document Goal ...
1028 0

推荐镜像

更多