Oracle與PostgreSQL的NULL和索引使用區(qū)別是什么

這篇文章主要講解了“Oracle與PostgreSQL的NULL和索引使用區(qū)別是什么”，文中的講解內(nèi)容簡單清晰，易于學(xué)習(xí)與理解，下面請大家跟著小編的思路慢慢深入，一起來研究和學(xué)習(xí)“Oracle與PostgreSQL的NULL和索引使用區(qū)別是什么”吧！

Oracle在創(chuàng)建索引時(shí),不會存儲NULL值,而PostgreSQL在創(chuàng)建索引時(shí)則會存儲NULL值.在查詢時(shí),如使用Column is null這樣的條件查詢,Oracle不會使用索引而PostgreSQL則會使用索引.

Oracle
插入數(shù)據(jù),200w多行的數(shù)據(jù),然后插入一行值為null的數(shù)據(jù).

TEST-orcl@DESKTOP-V430TU3>create table tbl1(id int);
Table created.
TEST-orcl@DESKTOP-V430TU3>create global temporary table tmp(id int);
Table created.
TEST-orcl@DESKTOP-V430TU3>insert into tmp select rownum from dba_objects;
133456 rows created.
TEST-orcl@DESKTOP-V430TU3>insert into tmp select * from tmp;
133455 rows created.
TEST-orcl@DESKTOP-V430TU3>/
266910 rows created.
TEST-orcl@DESKTOP-V430TU3>/
533820 rows created.
TEST-orcl@DESKTOP-V430TU3>/
1067640 rows created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 select * from tmp;
2135296 rows created.
TEST-orcl@DESKTOP-V430TU3>commit;
Commit complete.
TEST-orcl@DESKTOP-V430TU3>exec dbms_stats.gather_table_stats('TEST','TBL1',cascade=>true);
PL/SQL procedure successfully completed.
TEST-orcl@DESKTOP-V430TU3>select index_name,index_type,blevel,leaf_blocks,num_rows,status,distinct_keys from user_indexes  where table_name='TBL1';
INDEX_NAME                     INDEX_TYPE                      BLEVEL
------------------------------ --------------------------- ----------
LEAF_BLOCKS   NUM_ROWS STATUS   DISTINCT_KEYS
----------- ---------- -------- -------------
IDX_TBL1_ID                    NORMAL                               2
       4662    2103843 VALID           134688
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(null);
1 row created.
TEST-orcl@DESKTOP-V430TU3>commit;
Commit complete.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>exec dbms_stats.gather_table_stats('TEST','TBL1',cascade=>true);
PL/SQL procedure successfully completed.
TEST-orcl@DESKTOP-V430TU3>select index_name,index_type,blevel,leaf_blocks,num_rows,status,distinct_keys from user_indexes  where table_name='TBL1';
INDEX_NAME                     INDEX_TYPE                      BLEVEL
------------------------------ --------------------------- ----------
LEAF_BLOCKS   NUM_ROWS STATUS   DISTINCT_KEYS
----------- ---------- -------- -------------
IDX_TBL1_ID                    NORMAL                               2
       4771    2152683 VALID           134688

執(zhí)行查詢

TEST-orcl@DESKTOP-V430TU3>set autotrace on explain
TEST-orcl@DESKTOP-V430TU3>select * from tbl1 where id is null;
        ID
----------
Execution Plan
----------------------------------------------------------
Plan hash value: 312383637
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |     1 |     5 |   898   (2)| 00:00:11 |
|*  1 |  TABLE ACCESS FULL| TBL1 |     1 |     5 |   898   (2)| 00:00:11 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("ID" IS NULL)
TEST-orcl@DESKTOP-V430TU3>

PostgreSQL
數(shù)據(jù)表tbl1結(jié)構(gòu)與Oracle一致.

testdb=# insert into tbl1 select generate_series(1,100000);
INSERT 0 100000
testdb=# explain (analyze,verbose)  select * from tbl1 where id is null;
                                                 QUERY PLAN                                                 
--------------------------------------------------------------------------------------------------------
 Seq Scan on public.tbl1  (cost=0.00..1569.33 rows=11 width=4) (actual time=26.052..130.752 rows=1 loops=1)
   Output: id
   Filter: (tbl1.id IS NULL)
   Rows Removed by Filter: 110000
 Planning Time: 1.403 ms
 Execution Time: 130.814 ms
(6 rows)
testdb=# create index idx_tb1_id on tbl1(id);
CREATE INDEX
testdb=# explain (analyze,verbose)  select * from tbl1 where id is null;
                                                         QUERY PLAN                                                          
--------------------------------------------------------------------------------------------------------
 Index Only Scan using idx_tb1_id on public.tbl1  (cost=0.42..8.56 rows=4 width=4) (actual time=0.133..0.136 rows=1 loops=1)
   Output: id
   Index Cond: (tbl1.id IS NULL)
   Heap Fetches: 1
 Planning Time: 1.512 ms
 Execution Time: 0.199 ms
(6 rows)

使用id is null進(jìn)行查詢,使用的是Index Only Scan.

感謝各位的閱讀，以上就是“Oracle與PostgreSQL的NULL和索引使用區(qū)別是什么”的內(nèi)容了，經(jīng)過本文的學(xué)習(xí)后，相信大家對Oracle與PostgreSQL的NULL和索引使用區(qū)別是什么這一問題有了更深刻的體會，具體使用情況還需要大家實(shí)踐驗(yàn)證。這里是億速云，小編將為大家推送更多相關(guān)知識點(diǎn)的文章，歡迎關(guān)注！