zoukankan      html  css  js  c++  java
  • 搜集统计信息

    为了讲解直方图,我收集统计信息的时候是 method_opt => 'for all columns size skewonly'
    正式的生产环境中,最好别用allcolumns方式收集直方图,因为all columns 几乎会对所有列都收集直方图信息
    
    method_opt => 'for all columns size skewonly'
    
    drop table p500 purge;
    
    create table p500 as select * from dba_objects;
    
    
    
    select a.column_name,
              b.num_rows,
               a.num_distinct Cardinality,
               round(a.num_distinct / b.num_rows * 100, 2) selectivity,
              a.histogram,
              a.num_buckets
          from dba_tab_col_statistics a, dba_tables b
         where a.owner = b.owner
           and a.table_name = b.table_name
          and a.owner = 'TEST'
          and a.table_name = 'P500';
          
    
          
          select owner, table_name name, object_type, stale_stats, last_analyzed
      from dba_tab_statistics
     where table_name in ('P500')
       and owner = 'TEST'
       and (stale_stats = 'YES' or last_analyzed is null);
    
    
    
    
    BEGIN
          DBMS_STATS.GATHER_TABLE_STATS(ownname          => 'SCOTT',
                                        tabname          => 'TEST',
                                        estimate_percent => 100,
                                        method_opt       => 'for all columns size skewonly',
                                        no_invalidate    => FALSE,
                                        degree           => 1,
                                        cascade          => TRUE);
        END;
    
    
    
    method_opt => 'for all columns size auto',
    auto表示Oracle根据谓词过滤信息(前文讲解直方图的时候提到过的where条件过滤),自动判断该列是否收集直方图。
    
    一个稳定的系统,不应该让Oracle去自动判断,自动判断很可能就会出事,比如某列不该收集直方图,设置auto过后它自己去收集直方图了,从而导致系统不稳定。
    drop table p600 purge;
    
    create table p600 as select * from dba_objects;
    
    
        select a.column_name,
              b.num_rows,
               a.num_distinct Cardinality,
               round(a.num_distinct / b.num_rows * 100, 2) selectivity,
              a.histogram,
              a.num_buckets
          from dba_tab_col_statistics a, dba_tables b
         where a.owner = b.owner
           and a.table_name = b.table_name
          and a.owner = 'TEST'
          and a.table_name = 'P600';
          
    
    method_opt => 'for all columns size auto';
    
    select * from p600 a where a.owner='SYS';
    
    
    
    BEGIN
      DBMS_STATS.GATHER_TABLE_STATS(ownname          => 'TEST',
                                    tabname          => 'P600',
                                    estimate_percent => 100,
                                    method_opt       => 'for all columns size auto',
                                    no_invalidate    => FALSE,
                                    degree           => 8,
                                    cascade          => TRUE);
    END;
    
    
    
        select a.column_name,
              b.num_rows,
               a.num_distinct Cardinality,
               round(a.num_distinct / b.num_rows * 100, 2) selectivity,
              a.histogram,
              a.num_buckets
          from dba_tab_col_statistics a, dba_tables b
         where a.owner = b.owner
           and a.table_name = b.table_name
          and a.owner = 'TEST'
          and a.table_name = 'P600';
          
    
    只有OWNER列被采集了直方图
    
    SQL> set linesie 200
    SP2-0158: 未知的 SET 选项 "linesie"
    SQL> set linesize 200
    SQL> set pagesize 200
    SQL>     select a.column_name,
              b.num_rows,
               a.num_distinct Cardinality,
               round(a.num_distinct / b.num_rows * 100, 2) selectivity,
              a.histogram,
              a.num_buckets
          from dba_tab_col_statistics a, dba_tables b
         where a.owner = b.owner
           and a.table_name = b.table_name
          and a.owner = 'TEST'
          and a.table_name = 'P600';
            2    3    4    5    6    7    8    9   10   11  
    COLUMN_NAME										     NUM_ROWS CARDINALITY SELECTIVITY HISTOGRAM 				    NUM_BUCKETS
    ------------------------------------------------------------------------------------------ ---------- ----------- ----------- --------------------------------------------- -----------
    OWNER												86978	       30	  .03 FREQUENCY 					     30
    OBJECT_NAME											86978	    52436	60.29 NONE						      1
    SUBOBJECT_NAME											86978	      142	  .16 NONE						      1
    OBJECT_ID											86978	    86978	  100 NONE						      1
    DATA_OBJECT_ID											86978	     9097	10.46 NONE						      1
    OBJECT_TYPE											86978	       45	  .05 NONE						      1
    CREATED 											86978	      963	 1.11 NONE						      1
    LAST_DDL_TIME											86978	     1065	 1.22 NONE						      1
    TIMESTAMP											86978	     1108	 1.27 NONE						      1
    STATUS												86978		1	    0 NONE						      1
    TEMPORARY											86978		2	    0 NONE						      1
    GENERATED											86978		2	    0 NONE						      1
    SECONDARY											86978		2	    0 NONE						      1
    NAMESPACE											86978	       21	  .02 NONE						      1
    EDITION_NAME											86978		0	    0 NONE						      0
    
    已选择15行。
    
    
    
    method_opt=> 'for all columns size repeat'
    repeat表示以前收集过直方图,现在收集统计信息的时候就收集直方图,如果以前没收集过直方图,现在收集统计信息的时候就不收集。
    
    
    BEGIN
          DBMS_STATS.GATHER_TABLE_STATS(ownname          => 'TEST',
                                        tabname          => 'P500',
                                        estimate_percent => 100,
                                        
                                        method_opt=> 'for all columns size repeat',
                                        no_invalidate    => FALSE,
                                        degree           => 1,
                                        cascade          => TRUE);
        END;
    
    
    SQL> 
    SQL> 
    SQL>     select a.column_name,
              b.num_rows,
               a.num_distinct Cardinality,
               round(a.num_distinct / b.num_rows * 100, 2) selectivity,
              a.histogram,
              a.num_buckets
          from dba_tab_col_statistics a, dba_tables b
         where a.owner = b.owner
           and a.table_name = b.table_name
          and a.owner = 'TEST'
          and a.table_name = 'P600';  2    3    4    5    6    7    8    9   10   11  
    
    COLUMN_NAME										     NUM_ROWS CARDINALITY SELECTIVITY HISTOGRAM 				    NUM_BUCKETS
    ------------------------------------------------------------------------------------------ ---------- ----------- ----------- --------------------------------------------- -----------
    OWNER												86978	       30	  .03 FREQUENCY 					     30
    OBJECT_NAME											86978	    52436	60.29 NONE						      1
    SUBOBJECT_NAME											86978	      142	  .16 NONE						      1
    OBJECT_ID											86978	    86978	  100 NONE						      1
    DATA_OBJECT_ID											86978	     9097	10.46 NONE						      1
    OBJECT_TYPE											86978	       45	  .05 NONE						      1
    CREATED 											86978	      963	 1.11 NONE						      1
    LAST_DDL_TIME											86978	     1065	 1.22 NONE						      1
    TIMESTAMP											86978	     1108	 1.27 NONE						      1
    STATUS												86978		1	    0 NONE						      1
    TEMPORARY											86978		2	    0 NONE						      1
    GENERATED											86978		2	    0 NONE						      1
    SECONDARY											86978		2	    0 NONE						      1
    NAMESPACE											86978	       21	  .02 NONE						      1
    EDITION_NAME											86978		0	    0 NONE						      0
    
    已选择15行。
    
    
  • 相关阅读:
    IOS GameCenter验证登陆
    BZOJ 1597: [Usaco2008 Mar]土地购买 斜率优化
    Hihocoder #1602 : 本质不同的回文子串的数量 manacher + BKDRhash
    HDU 5343 MZL's Circle Zhou 后缀自动机+DP
    HDU 6208 The Dominator of Strings 后缀自动机
    SPOJ SUBLEX
    SPOJ LCS2
    .NET平台技术体系梳理+初学者学习路径推荐+我们的愿景与目标
    扑克模拟,牌型判断java版
    一道综合练习题实践list及dictionary集合类
  • 原文地址:https://www.cnblogs.com/hzcya1995/p/13348773.html
Copyright © 2011-2022 走看看