6.没有对表查询前还是未真正放到In-Memory内存中
SQL> SELECT * FROM V$INMEMORY_AREA;
POOL                 ALLOC_BYTES USED_BYTES POPULATE_STATUS          CON_ID
-------------------- ----------- ---------- -------------------- ----------
1MB POOL               837812224          0 DONE                          3
64KB POOL              201326592          0 DONE                          3
SQL> select count(*) from bmw.t;
COUNT(*)
----------
90927
　　7.查询一次后可以看到USED_BYTES已经分配了内存
　　SQL> SELECT * FROM V$INMEMORY_AREA;
　　POOL                 ALLOC_BYTES USED_BYTES POPULATE_STATUS          CON_ID
　　-------------------- ----------- ---------- -------------------- ----------
　　1MB POOL               837812224    4194304 DONE                          3
　　64KB POOL              201326592     131072 DONE                          3
　　8.查看一下执行计划
SQL> set autot trace
SQL>  SELECT * FROM bmw.t;
90927 rows selected.
Execution Plan
----------------------------------------------------------
Plan hash value: 1601196873
-----------------------------------------------------------------------------------
| Id  | Operation                  | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------------
|   0 | SELECT STATEMENT           |      | 90927 |     9M|    20  (20)| 00:00:01 |
|   1 |  TABLE ACCESS INMEMORY FULL| T    | 90927 |     9M|    20  (20)| 00:00:01 |
-----------------------------------------------------------------------------------
Statistics
----------------------------------------------------------
5  recursive calls
0  db block gets
5  consistent gets
0  physical reads
0  redo size
4944325  bytes sent via SQL*Net to client
67223  bytes received via SQL*Net from client
6063  SQL*Net roundtrips to/from client
0  sorts (memory)
0  sorts (disk)
90927  rows processed
　　这里可以看到使用了TABLE ACCESS INMEMORY FULL扫描，而且consistent gets为5.可以看到INMEMORY会给查询带来非常高的性能提升。