如果 shared pool 设置过小，那么 data dictionary cache 也会相应的过小，没有足够的空间存储ORACLE的系统数据字典信息，从而导致ORACLE需要从硬盘读取数据字典信息，我们查看执行计划的时候看到的recursive calls就是代表的是从磁盘读取数据字典的次数。
SQL> select count(*) from dba_source;
COUNT(*)
----------
228222
Statistics
----------------------------------------------------------
312 recursive calls
0 db block gets
888 consistent gets
538 physical reads
0 redo size
413 bytes sent via SQL*Net to client
384 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
5 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(*) from dba_source;
COUNT(*)
----------
228222
Statistics
----------------------------------------------------------
0 recursive calls
0 db block gets
818 consistent gets
0 physical reads
0 redo size
413 bytes sent via SQL*Net to client
384 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
可以看到第一次没有访问数据字典，所以recursive calls等于312，第二次，由于缓存了数据字典信息，所以recursive calls为0 data
查看 data dictionary cache 的命中率
SQL> select sum(gets),sum(getmisses),(1-(sum(getmisses)/(sum(gets)+sum(getmisses)))) hitratio from v$rowcache;
SUM(GETS) SUM(GETMISSES) HITRATIO
---------- -------------- ----------
264014 15600 .944208802
查看data dictionary cache 的大小
SQL> select sum(sharable_mem) from v$sqlarea;
SUM(SHARABLE_MEM)
-----------------
14354029
如果data dictionary cache的命中率小于95%，应该适当增加shared pool 的大小