一、什么是執(zhí)行計(jì)劃
An explain plan is a representation of the access path that is taken when a query is executed within Oracle.
二、如何訪(fǎng)問(wèn)數(shù)據(jù)
At the physical level Oracle reads blocks of data. The smallest amount of data read is a single Oracle block, the largest is constrained by operating system limits (and multiblock i/o). Logically Oracle finds the data to read by using the following methods:
Full Table Scan (FTS) --全表掃描
Index Lookup (unique & non-unique) --索引掃描(唯一和非唯一)
Rowid --物理行id
三、執(zhí)行計(jì)劃層次關(guān)系
When looking at a plan, the rightmost (ie most inndented) uppermost operation is the first thing that is executed. --采用最右最上最先執(zhí)行的原則看層次關(guān)系,在同一級(jí)如果某個(gè)動(dòng)作沒(méi)有子ID就最先執(zhí)行
1.一個(gè)簡(jiǎn)單的例子:
SQL> select /*+parallel (e 4)*/ * from emp e;
Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=CHOOSE (Cost=1 Card=82 Bytes=7134)
1 0 TABLE ACCESS* (FULL) OF 'EMP' (Cost=1 Card=82 Bytes=7134):Q5000
--[:Q5000]表示是并行方式
1 PARALLEL_TO_SERIAL SELECT /*+ NO_EXPAND ROWID(A1) */ A1."EMPNO"
,A1."ENAME",A1."JOB",A1."MGR",A1."HI
優(yōu)化模式是CHOOSE的情況下,看Cost參數(shù)是否有值來(lái)決定采用CBO還是RBO:
SELECT STATEMENT [CHOOSE] Cost=1234 --Cost有值,采用CBO
SELECT STATEMENT [CHOOSE] --Cost為空,采用RBO(9I是如此顯示的)
2.層次的父子關(guān)系的例子:
PARENT1
**FIRST CHILD
****FIRST GRANDCHILD
**SECOND CHILD
Here the same principles apply, the FIRST GRANDCHILD is the initial operation then the FIRST CHILD followed by the SECOND CHILD and finally the PARENT collates the output.
四、例子解說(shuō)
Execution Plan
----------------------------------------------------------
0 **SELECT STATEMENT Optimizer=CHOOSE (Cost=3 Card=8 Bytes=248)
1 0 **HASH JOIN (Cost=3 Card=8 Bytes=248)
2 1 ****TABLE ACCESS (FULL) OF 'DEPT' (Cost=1 Card=3 Bytes=36)
3 1 ****TABLE ACCESS (FULL) OF 'EMP' (Cost=1 Card=16 Bytes=304)
左側(cè)的兩排數(shù)據(jù),前面的是序列號(hào)ID,后面的是對(duì)應(yīng)的PID(父ID)。
A shortened summary of this is:
Execution starts with ID=0: SELECT STATEMENT but this is dependand on it's child objects
So it executes its first child step: ID=1 PID=0 HASH JOIN but this is dependand on it's child objects
So it executes its first child step: ID=2 PID=1 TABLE ACCESS (FULL) OF 'DEPT'
Then the second child step: ID=3 PID=2 TABLE ACCESS (FULL) OF 'EMP'
Rows are returned to the parent step(s) until finished
五、表訪(fǎng)問(wèn)方式
1.Full Table Scan (FTS) 全表掃描
In a FTS operation, the whole table is read up to the high water mark (HWM). The HWM marks the last block in the table that has ever had data written to it. If you have deleted all the rows then you will still read up to the HWM. Truncate resets the HWM back to the start of the table. FTS uses multiblock i/o to read the blocks from disk. --全表掃描模式下會(huì)讀數(shù)據(jù)到表的高水位線(xiàn)(HWM即表示表曾經(jīng)擴(kuò)展的最后一個(gè)數(shù)據(jù)塊),讀取速度依賴(lài)于Oracle初始化參數(shù)db_block_multiblock_read_count(我覺(jué)得應(yīng)該這樣翻譯:FTS掃描會(huì)使表使用上升到高水位(HWM),HWM標(biāo)識(shí)了表最后寫(xiě)入數(shù)據(jù)的塊,如果你用DELETE刪除了所有的數(shù)據(jù)表仍然處于高水位(HWM),只有用TRUNCATE才能使表回歸,FTS使用多IO從磁盤(pán)讀取數(shù)據(jù)塊).
Query Plan
------------------------------------
SELECT STATEMENT [CHOOSE] Cost=1
**INDEX UNIQUE SCAN EMP_I1 --如果索引里就找到了所要的數(shù)據(jù),就不會(huì)再去訪(fǎng)問(wèn)表
2.Index Lookup 索引掃描
There are 5 methods of index lookup:
index unique scan --索引唯一掃描
Method for looking up a single key value via a unique index. always returns a single value, You must supply AT LEAST the leading column of the index to access data via the index.
eg:SQL> explain plan for select empno,ename from emp where empno=10;
index range scan --索引局部掃描
Index range scan is a method for accessing a range values of a particular column. AT LEAST the leading column of the index must be supplied to access data via the index. Can be used for range operations (e.g. > < <> >= <= between) .
eg:SQL> explain plan for select mgr from emp where mgr = 5;
index full scan --索引全局掃描
Full index scans are only available in the CBO as otherwise we are unable to determine whether a full scan would be a good idea or not. We choose an index Full Scan when we have statistics that indicate that it is going to be more efficient than a Full table scan and a sort. For example we may do a Full index scan when we do an unbounded scan of an index and want the data to be ordered in the index order.
eg: SQL> explain plan for select empno,ename from big_emp order by empno,ename;
index fast full scan --索引快速全局掃描,不帶order by情況下常發(fā)生
Scans all the block in the index, Rows are not returned in sorted order, Introduced in 7.3 and requires V733_PLANS_ENABLED=TRUE and CBO, may be hinted using INDEX_FFS hint, uses multiblock i/o, can be executed in parallel, can be used to access second column of concatenated indexes. This is because we are selecting all of the index.
eg: SQL> explain plan for select empno,ename from big_emp;
index skip scan --索引跳躍掃描,where條件列是非索引的前導(dǎo)列情況下常發(fā)生
Index skip scan finds rows even if the column is not the leading column of a concatenated index. It skips the first column(s) during the search.
eg:SQL> create index i_emp on emp(empno, ename);
SQL> select /*+ index_ss(emp i_emp)*/ job from emp where ename='SMITH';
3.Rowid 物理ID掃描
This is the quickest access method available.Oracle retrieves the specified block and extracts the rows it is interested in. --Rowid掃描是最快的訪(fǎng)問(wèn)數(shù)據(jù)方式
六、表連接方式
七、運(yùn)算符
1.sort --排序,很消耗資源
There are a number of different operations that promote sorts:
(1)order by clauses (2)group by (3)sort merge join –-這三個(gè)會(huì)產(chǎn)生排序運(yùn)算
2.filter --過(guò)濾,如not in、min函數(shù)等容易產(chǎn)生
Has a number of different meanings, used to indicate partition elimination, may also indicate an actual filter step where one row source is filtering, another, functions such as min may introduce filter steps into query plans.
3.view --視圖,大都由內(nèi)聯(lián)視圖產(chǎn)生(可能深入到視圖基表)
When a view cannot be merged into the main query you will often see a projection view operation. This indicates that the 'view' will be selected from directly as opposed to being broken down into joins on the base tables. A number of constructs make a view non mergeable. Inline views are also non mergeable.
eg: SQL> explain plan for
select ename,tot from emp,(select empno,sum(empno) tot from big_emp group by empno) tmp
where emp.empno = tmp.empno;
Query Plan
------------------------
SELECT STATEMENT [CHOOSE]
**HASH JOIN
**TABLE ACCESS FULL EMP [ANALYZED]
**VIEW
****SORT GROUP BY
******INDEX FULL SCAN BE_IX
4.partition view --分區(qū)視圖
Partition views are a legacy technology that were superceded by the partitioning option. This section of the article is provided as reference for such legacy systems.
示例:假定A、B、C都是不是小表,且在A表上一個(gè)組合索引:A(a.col1,a.col2) ,注意a.col1列為索引的引導(dǎo)列。考慮下面的查詢(xún):
select A.col4 from A , B , C
where B.col3 = 10 and A.col1 = B.col1 and A.col2 = C.col2 and C.col3 = 5;
Execution Plan
------------------------------------
0 SELECT STATEMENT Optimizer=CHOOSE
1 0 MERGE JOIN
2 1 SORT (JOIN)
3 2 NESTED LOOPS
4 3 TABLE ACCESS (FULL) OF 'B'
5 3 TABLE ACCESS (BY INDEX ROWID) OF 'A'
6 5 INDEX (RANGE SCAN) OF 'INX_COL12A' (NON-UNIQUE)
7 1 SORT (JOIN)
8 7 TABLE ACCESS (FULL) OF 'C'
Statistics(統(tǒng)計(jì)信息參數(shù),參見(jiàn)另外個(gè)轉(zhuǎn)載的文章)
--------------------------------------
0 recursive calls(歸調(diào)用次數(shù))
8 db block gets(從磁盤(pán)上讀取的塊數(shù),即通過(guò)update/delete/select for update讀的次數(shù))
6 consistent gets(從內(nèi)存里讀取的塊數(shù),即通過(guò)不帶for update的select 讀的次數(shù))
0 physical reads(物理讀—從磁盤(pán)讀到數(shù)據(jù)塊數(shù)量,一般來(lái)說(shuō)是'consistent gets' + 'db block gets')
0 redo size (重做數(shù)——執(zhí)行SQL的過(guò)程中,產(chǎn)生的重做日志的大小)
551 bytes sent via SQL*Net to client
430 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
2 sorts (memory) (在內(nèi)存中發(fā)生的排序)
0 sorts (disk) (在硬盤(pán)中發(fā)生的排序)
6 rows processed
在表做連接時(shí),只能2個(gè)表先做連接,然后將連接后的結(jié)果作為一個(gè)row source,與剩下的表做連接,在上面的例子中,連接順序?yàn)锽與A先連接,然后再與C連接:
B <---> A <---> C
col3=10 col3=5
如果沒(méi)有執(zhí)行計(jì)劃,分析一下,上面的3個(gè)表應(yīng)該拿哪一個(gè)作為第一個(gè)驅(qū)動(dòng)表?從SQL語(yǔ)句看來(lái),只有B表與C表上有限制條件,所以第一個(gè)驅(qū)動(dòng)表應(yīng)該為這2個(gè)表中的一個(gè),到底是哪一個(gè)呢?
B表有謂詞B.col3 = 10,這樣在對(duì)B表做全表掃描的時(shí)候就將where子句中的限制條件(B.col3 = 10)用上,從而得到一個(gè)較小的row source, 所以B表應(yīng)該作為第一個(gè)驅(qū)動(dòng)表。而且這樣的話(huà),如果再與A表做關(guān)聯(lián),可以有效利用A表的索引(因?yàn)锳表的col1列為leading column)。
上面的查詢(xún)中C表上也有謂詞(C.col3 = 5),有人可能認(rèn)為C表作為第一個(gè)驅(qū)動(dòng)表也能獲得較好的性能。讓我們?cè)賮?lái)分析一下:如果C表作為第一個(gè)驅(qū)動(dòng)表,則能保證驅(qū)動(dòng)表生成很小的row source,但是看看連接條件A.col2 = C.col2,此時(shí)就沒(méi)有機(jī)會(huì)利用A表的索引,因?yàn)锳表的col2列不為leading column,這樣nested loop的效率很差,從而導(dǎo)致查詢(xún)的效率很差。所以對(duì)于NL連接選擇正確的驅(qū)動(dòng)表很重要。
因此上面查詢(xún)比較好的連接順序?yàn)?B - - > A) - - > C。如果數(shù)據(jù)庫(kù)是基于代價(jià)的優(yōu)化器,它會(huì)利用計(jì)算出的代價(jià)來(lái)決定合適的驅(qū)動(dòng)表與合適的連接順序。一般來(lái)說(shuō),CBO都會(huì)選擇正確的連接順序,如果CBO選擇了比較差的連接順序,我們還可以使用Oracle提供的hints來(lái)讓CBO采用正確的連接順序。如下所示
select /*+ ordered */ A.col4
from B,A,C
where B.col3 = 10 and A.col1 = B.col1 and A.col2 = C.col2 and C.col3 = 5
既然選擇正確的驅(qū)動(dòng)表這么重要,那么讓我們來(lái)看一下執(zhí)行計(jì)劃,到底各個(gè)表之間是如何關(guān)聯(lián)的,從而得到執(zhí)行計(jì)劃中哪個(gè)表應(yīng)該為驅(qū)動(dòng)表:
在執(zhí)行計(jì)劃中,需要知道哪個(gè)操作是先執(zhí)行的,哪個(gè)操作是后執(zhí)行的,這對(duì)于判斷哪個(gè)表為驅(qū)動(dòng)表有用處。判斷之前,如果對(duì)表的訪(fǎng)問(wèn)是通過(guò)rowid,且該rowid的值是從索引掃描中得來(lái)得,則將該索引掃描先從執(zhí)行計(jì)劃中暫時(shí)去掉。然后在執(zhí)行計(jì)劃剩下的部分中,判斷執(zhí)行順序的指導(dǎo)原則就是:最右、最上的操作先執(zhí)行。具體解釋如下:
得到去除妨礙判斷的索引掃描后的執(zhí)行計(jì)劃:Execution Plan
-------------------------------------
0 SELECT STATEMENT Optimizer=CHOOSE
1 0 MERGE JOIN
2 1 SORT (JOIN)
3 2 NESTED LOOPS
4 3 TABLE ACCESS (FULL) OF 'B'
5 3 TABLE ACCESS (BY INDEX ROWID) OF 'A'
6 5 INDEX (RANGE SCAN) OF 'INX_COL12A' (NON-UNIQUE)
7 1 SORT (JOIN)
8 7 TABLE ACCESS (FULL) OF 'C'
看執(zhí)行計(jì)劃的第3列,即字母部分,每列值的左面有空格作為縮進(jìn)字符。在該列值左邊的空格越多,說(shuō)明該列值的縮進(jìn)越多,該列值也越靠右。如上面的執(zhí)行計(jì)劃所示:第一列值為6的行的縮進(jìn)最多,即該行最靠右;第一列值為4、5的行的縮進(jìn)一樣,其靠右的程度也一樣,但是第一列值為4的行比第一列值為5的行靠上;談?wù)撋舷玛P(guān)系時(shí),只對(duì)連續(xù)的、縮進(jìn)一致的行有效。
從這個(gè)圖中我們可以看到,對(duì)于NESTED LOOPS部分,最右、最上的操作是TABLE ACCESS (FULL) OF 'B',所以這一操作先執(zhí)行,所以該操作對(duì)應(yīng)的B表為第一個(gè)驅(qū)動(dòng)表(外部表),自然,A表就為內(nèi)部表了。從圖中還可以看出,B與A表做嵌套循環(huán)后生成了新的row source ,對(duì)該row source進(jìn)行來(lái)排序后,與C表對(duì)應(yīng)的排序了的row source(應(yīng)用了C.col3 = 5限制條件)進(jìn)行SMJ連接操作。所以從上面可以得出如下事實(shí):B表先與A表做嵌套循環(huán),然后將生成的row source與C表做排序—合并連接。
通過(guò)分析上面的執(zhí)行計(jì)劃,我們不能說(shuō)C表一定在B、A表之后才被讀取,事實(shí)上,B表有可能與C表同時(shí)被讀入內(nèi)存,因?yàn)閷⒈碇械臄?shù)據(jù)讀入內(nèi)存的操作可能為并行的。事實(shí)上許多操作可能為交叉進(jìn)行的,因?yàn)镺racle讀取數(shù)據(jù)時(shí),如果就是需要一行數(shù)據(jù)也是將該行所在的整個(gè)數(shù)據(jù)塊讀入內(nèi)存,而且有可能為多塊讀。
看執(zhí)行計(jì)劃時(shí),我們的關(guān)鍵不是看哪個(gè)操作先執(zhí)行,哪個(gè)操作后執(zhí)行,而是關(guān)鍵看表之間連接的順序(如得知哪個(gè)為驅(qū)動(dòng)表,這需要從操作的順序進(jìn)行判斷)、使用了何種類(lèi)型的關(guān)聯(lián)及具體的存取路徑(如判斷是否利用了索引)
在從執(zhí)行計(jì)劃中判斷出哪個(gè)表為驅(qū)動(dòng)表后,根據(jù)我們的知識(shí)判斷該表作為驅(qū)動(dòng)表(就像上面判斷ABC表那樣)是否合適,如果不合適,對(duì)SQL語(yǔ)句進(jìn)行更改,使優(yōu)化器可以選擇正確的驅(qū)動(dòng)表。 本文出自:億恩科技【1tcdy.com】
服務(wù)器租用/服務(wù)器托管中國(guó)五強(qiáng)!虛擬主機(jī)域名注冊(cè)頂級(jí)提供商!15年品質(zhì)保障!--億恩科技[ENKJ.COM]
|