set autotrace traceonly explain
select * from emp where deptno=20;
Execution Plan
----------------------------------------------------------
Plan hash value: 2468466201
----------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 5 | 160 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 5 | 160 | 2 (0)| 00:00:01 |
|* 2 | INDEX RANGE SCAN | EMP_DEPTNO_IDX | 5 | | 1 (0)| 00:00:01 |
----------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("DEPTNO"=20)
create index emp_idx on emp(ename,sal);
인덱스가 생성되었습니다.
1. 인덱스 첫 번째 컬럼이 조건절에서 사용되지 않았으나 ename으로 정렬해야 하므로 emp_idx 인덱스 사용하는 것이 Table Full Scan보다 유리
1) INDEX FULL SCAN 선택
set autotrace traceonly explain
select * from emp where sal > 2000 order by ename;
Execution Plan
----------------------------------------------------------
Plan hash value: 737262432
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10 | 320 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 10 | 320 | 2 (0)| 00:00:01 |
|* 2 | INDEX FULL SCAN | EMP_IDX | 10 | | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("SAL">2000)
filter("SAL">2000)
2) TABLE ACCESS FULL 선택 : order by ename이 빠짐
set autotrace traceonly explain
select * from emp where sal > 2000;
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10 | 320 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 10 | 320 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("SAL">2000)
2. 인덱스를 이용한 소트 연산 대체
1) 책에서는 first_rows 힌트의 사용으로인해 옵티마이저는 소트연산을 생략함으로써 전체 집합 중 처음 일부만을 빠르게 리턴할 목적으로
Index Full Scan방식을 선택 했다고 하나, 실제 테스트에서는 힌트의 사용 유무와 상관없이 Index Full Scan방식이 선택됨
※ 각 단계별로 "alter system flush buffer_cache; & alter system flush shared_pool;"를 수행 시에도 아래와 동일한 결과가 나옴
set autotrace traceonly explain
select /*+ first_rows */ * from emp where sal > 1000 order by ename;
Execution Plan
----------------------------------------------------------
Plan hash value: 737262432
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 13 | 416 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 13 | 416 | 2 (0)| 00:00:01 |
|* 2 | INDEX FULL SCAN | EMP_IDX | 13 | | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("SAL">1000)
filter("SAL">1000)
set autotrace traceonly explain
select * from emp where sal > 1000 order by ename;
Execution Plan
----------------------------------------------------------
Plan hash value: 737262432
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 13 | 416 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 13 | 416 | 2 (0)| 00:00:01 |
|* 2 | INDEX FULL SCAN | EMP_IDX | 13 | | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("SAL">1000)
filter("SAL">1000)
set autotrace traceonly explain
select * from emp where sal > 1000;
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 13 | 416 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 13 | 416 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("SAL">1000)
set autotrace traceonly explain
select /*+ first_rows */ * from emp where sal > 1000;
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 13 | 416 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 13 | 416 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("SAL">1000)
set autotrace traceonly explain
select empno,ename from emp where empno=7788;
Execution Plan
----------------------------------------------------------
Plan hash value: 2949544139
--------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 8 | 1 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 1 | 8 | 1 (0)| 00:00:01 |
|* 2 | INDEX UNIQUE SCAN | PK_EMP | 1 | | 0 (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("EMPNO"=7788)
index skip scan과 in-list iterator의 동작 원리
set autotrace traceonly explain
select * from emp where sal between 2000 and 4000;
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 9 | 288 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 9 | 288 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("SAL">=2000 AND "SAL"<=4000)
set autotrace traceonly explain
select /*+ index_ss(emp emp_idx) */ * from emp where sal between 2000 and 4000;
Execution Plan
----------------------------------------------------------
Plan hash value: 302813946
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 9 | 288 | 15 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP | 9 | 288 | 15 (0)| 00:00:01 |
|* 2 | INDEX SKIP SCAN | EMP_IDX | 7 | | 14 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("SAL">=2000 AND "SAL"<=4000)
filter("SAL">=2000 AND "SAL"<=4000)
set autotrace traceonly explain
select * from emp where sal between 2000 and 4000 and job in ('MANAGER','ANALYST');
Execution Plan
----------------------------------------------------------
Plan hash value: 1754328461
--------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 128 | 2 (0)| 00:00:01 |
| 1 | INLIST ITERATOR | | | | | |
|* 2 | TABLE ACCESS BY INDEX ROWID| EMP | 4 | 128 | 2 (0)| 00:00:01 |
|* 3 | INDEX RANGE SCAN | EMP_JOB_IDX | 6 | | 1 (0)| 00:00:01 |
--------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("SAL">=2000 AND "SAL"<=4000)
3 - access("JOB"='ANALYST' OR "JOB"='MANAGER')
Index Full Scan | Index Fast Full Scan |
---|---|
1. 인덱스 구조를 따랄 스캔 2. 결과 집합 순서 보장 3. Single Block I/O 4. 병렬스캔 불가(파티션 되어 있지 않을 경우) 5. 인덱스에 포함되지 않은 컬럼 조회 시에도 사용 가능 | 1. 세그먼트 전체를 스캔 2. 결과집합 순서 보장 안 됨 3. multiblock I/O 4. 병렬스캔 가능 5. 인덱스에 포함된 컬럼으로만 조회할 때 사용 가능 |
SQL> conn sys/loveora as sysdba
Connected.
SQL> show parameter multiblock
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_file_multiblock_read_count integer 16
SQL> conn scott/loveora
Connected.
SQL> create index big_table_idx on big_table(object_name);
Index created.
alter system flush buffer_cache;
alter system flush shared_pool;
1. Index Full scan
ALTER SESSION SET EVENTS '10046 trace name context forever, level 8';
select /*+ index(b big_table_idx) */ count(object_name) from big_table b;
COUNT(OBJECT_NAME)
------------------
10000000
call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ---------- ----------
Parse 2 0.01 0.02 0 0 0 0
Execute 2 0.00 0.00 0 0 0 0
Fetch 4 21.27 33.84 101164 101166 0 2
------- ------ -------- ---------- ---------- ---------- ---------- ----------
total 8 21.29 33.87 101164 101166 0 2
Misses in library cache during parse: 1
Optimizer mode: ALL_ROWS
Parsing user id: 54 (SCOTT)
Rows Row Source Operation
------- ---------------------------------------------------
1 SORT AGGREGATE (cr=50583 pr=50582 pw=0 time=17442436 us)
10000000 INDEX FULL SCAN BIG_TABLE_IDX (cr=50583 pr=50582 pw=0 time=370001596 us)(object id 53503)
Rows Execution Plan
------- ---------------------------------------------------
0 SELECT STATEMENT MODE: ALL_ROWS
1 SORT (AGGREGATE)
10000000 INDEX MODE: ANALYZED (FULL SCAN) OF 'BIG_TABLE_IDX' (INDEX)
Elapsed times include waiting on following events:
Event waited on Times Max. Wait Total Waited
---------------------------------------- Waited ---------- ------------
SQL*Net message to client 4 0.00 0.00
db file sequential read 101164 0.00 5.52
SQL*Net message from client 4 0.00 0.00
1) INDEX FULL SCAN 시 db file sequential read는 Disk I/O와 동일 하게 일어남
2) db file sequential read : single block read 방식으로 I/O 할 때 발생되는 대기 이벤트
2. Index Fast Full Scan
select /*+ index_ffs(b big_table_idx) */ count(object_name) from big_table b;
COUNT(OBJECT_NAME)
------------------
10000000
call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ---------- ----------
Parse 1 0.00 0.00 0 0 0 0
Execute 1 0.00 0.00 0 0 0 0
Fetch 2 5.73 9.52 50817 50846 0 1
------- ------ -------- ---------- ---------- ---------- ---------- ----------
total 4 5.74 9.52 50817 50846 0 1
Misses in library cache during parse: 1
Optimizer mode: ALL_ROWS
Parsing user id: 54 (SCOTT)
Rows Row Source Operation
------- ---------------------------------------------------
1 SORT AGGREGATE (cr=50846 pr=50817 pw=0 time=9525459 us)
10000000 INDEX FAST FULL SCAN BIG_TABLE_IDX (cr=50846 pr=50817 pw=0 time=380001430 us)(object id 53503)
Rows Execution Plan
------- ---------------------------------------------------
0 SELECT STATEMENT MODE: ALL_ROWS
1 SORT (AGGREGATE)
10000000 INDEX MODE: ANALYZED (FAST FULL SCAN) OF 'BIG_TABLE_IDX'
(INDEX)
Elapsed times include waiting on following events:
Event waited on Times Max. Wait Total Waited
---------------------------------------- Waited ---------- ------------
SQL*Net message to client 2 0.00 0.00
db file scattered read 3204 0.00 0.73
db file sequential read 1 0.00 0.00
SQL*Net message from client 2 0.00 0.00
1) INDEX FAST FULL SCAN 시 db file scattered read는 3204회, Disk I/O는 50817 발생
2) 한번 디스크 I/O Call이 발생할 때마다 평균 15.86개의 블록을 퍼올려 읽음(db_file_multiblock_read_count = 16)
3) db file scattered read : multi block read 방식으로 I/O 할 때 발생되는 대기 이벤트
alter session set sql_trace=false;
set autotrace traceonly explain
select /*+ and_equal(e emp_deptno_idx emp_job_idx) */ * from emp e where deptno=30 and job='SALESMAN';
Execution Plan
----------------------------------------------------------
Plan hash value: 2839249562
----------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 32 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS BY INDEX ROWID| EMP | 1 | 32 | 3 (0)| 00:00:01 |
| 2 | AND-EQUAL | | | | | |
|* 3 | INDEX RANGE SCAN | EMP_JOB_IDX | 3 | | 1 (0)| 00:00:01 |
|* 4 | INDEX RANGE SCAN | EMP_DEPTNO_IDX | 5 | | 1 (0)| 00:00:01 |
----------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("JOB"='SALESMAN' AND "DEPTNO"=30)
3 - access("JOB"='SALESMAN')
4 - access("DEPTNO"=30)
set autotrace traceonly explain
select /*+ index_combine(e emp_deptno_idx emp_job_idx) */ * from emp e where deptno=30 and job='SALESMAN';
Execution Plan
----------------------------------------------------------
Plan hash value: 2413831903
---------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 32 | 4 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID | EMP | 1 | 32 | 4 (0)| 00:00:01 |
| 2 | BITMAP CONVERSION TO ROWIDS | | | | | |
| 3 | BITMAP AND | | | | | |
| 4 | BITMAP CONVERSION FROM ROWIDS| | | | | |
|* 5 | INDEX RANGE SCAN | EMP_JOB_IDX | | | 1 (0)| 00:00:01 |
| 6 | BITMAP CONVERSION FROM ROWIDS| | | | | |
|* 7 | INDEX RANGE SCAN | EMP_DEPTNO_IDX | | | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
5 - access("JOB"='SALESMAN')
7 - access("DEPTNO"=30)
1단계 : 일반 B*Tree 인덱스를 스캔하면서 각 조건에 만족하는 레코드의 rowid 목록 얻기(INDEX RANGE SCAN)
2단계 : 1단계에서 얻은 rowid목록을 가지고 비트맵 인덱스 구조 만들기(BITMAP CONVERSION FROM ROWIDS)
3단계 : 비트맵 인덱스에 대한 bit-wise operation 수행(BITMAP AND)
4단계 : bit-wise operation 수행 결과가 true인 비트 값들을 rowid 값으로 환산해 최종적으로 방문할 테이블 rowid 목록 얻기(BITMAP CONVERSION TO ROWIDS)
5단계 : rowid를 이용해 테이블 액세스(TABLE ACCESS BY INDEX ROWID)
set autotrace traceonly explain
select /*+ index_join(e emp_deptno_idx emp_job_idx) */ * from emp e where deptno=30 and job='SALESMAN';
Execution Plan
----------------------------------------------------------
Plan hash value: 2211876416
-------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 32 | 2 (0)| 00:00:01 |
|* 1 | TABLE ACCESS BY INDEX ROWID| EMP | 1 | 32 | 2 (0)| 00:00:01 |
|* 2 | INDEX RANGE SCAN | EMP_JOB_IDX | 3 | | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("DEPTNO"=30)
2 - access("JOB"='SALESMAN')
set autotrace traceonly explain
select /*+ index_join(e emp_deptno_idx emp_job_idx) */ deptno,job from emp e where deptno=30 and job='SALESMAN';
Execution Plan
----------------------------------------------------------
Plan hash value: 3557895725
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 9 | 3 (34)| 00:00:01 |
|* 1 | VIEW | index$_join$_001 | 1 | 9 | 3 (34)| 00:00:01 |
|* 2 | HASH JOIN | | | | | |
|* 3 | INDEX RANGE SCAN| EMP_JOB_IDX | 1 | 9 | 1 (0)| 00:00:01 |
|* 4 | INDEX RANGE SCAN| EMP_DEPTNO_IDX | 1 | 9 | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("JOB"='SALESMAN' AND "DEPTNO"=30)
2 - access(ROWID=ROWID)
3 - access("JOB"='SALESMAN')
4 - access("DEPTNO"=30)
1단계 : 크기가 비교적 작은 쪽 인덱스(emp_job_idx)에서 키 값과 rowid를 얻어 PGA 메모리에 해시 맵 생성
(해시 키 = rowid 가 사용)
2단계 : 다른 쪽 인덱스(emp_dept_idx)를 스캔하면서 먼저 생성한 해시 맵에 같은 rowid 값을 갖는 레코드가 있는지 탐색
3단계 : rowid끼리 조인에 성공한 레코드만 결과집합에 포함 : 이 rowid가 가리키는 테이블은 각 인덱스 컬럼에 대한 검색 조건을 모두 만족한다