網(wǎng)站首頁 編程語言 正文
最近碰到這樣一個(gè)SQL引發(fā)的性能問題,SQL內(nèi)容大致如下:
SELECT *
FROM t1
WHERE id = 999
AND (case $1
WHEN 'true' THEN
info = $2
ELSE info = $3 end) limit 1;
開發(fā)反應(yīng)這條SQL加上limit 1之后過了一段時(shí)間從原先的索引掃描變成了全表掃描,一個(gè)簡單的limit 1為何會(huì)產(chǎn)生這樣的影響,我只取一條數(shù)據(jù)不是應(yīng)該更快了嗎?
下面我們就從這條SQL開始說起。
首先我們先看下這個(gè)表結(jié)構(gòu),比較簡單,info列上有個(gè)索引,如下所示:
bill=# \d t1
Table "public.t1"
Column | Type | Collation | Nullable | Default
----------+-----------------------------+-----------+----------+---------
id | integer | | |
info | text | | |
crt_time | timestamp without time zone | | |
Indexes:
"idx_t1" btree (info)
并且info列是沒有重復(fù)值的,這意味著無論where條件中傳入什么變量都肯定是能走索引掃描的。那為什么加上limit 1后會(huì)變成全表掃描呢?
我們先看看這條SQL之前正常的走索引的執(zhí)行計(jì)劃:
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.027..0.027 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.025..0.026 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.158 ms
Execution Time: 0.057 ms
(6 rows)
而現(xiàn)在的執(zhí)行計(jì)劃卻是這樣的:
Limit (cost=0.00..0.35 rows=1 width=45) (actual time=487.564..487.564 rows=0 loops=1)
-> Seq Scan on t1 (cost=0.00..170895.98 rows=491791 width=45) (actual time=487.562..487.562 rows=0 loops=1)
Filter: ((id = 999) AND CASE $1 WHEN 'true'::text THEN (info = $2) ELSE (info = $3) END)
Rows Removed by Filter: 6000000
Planning Time: 0.119 ms
Execution Time: 487.595 ms
(6 rows)
奇怪的是下面的全表掃描加上limit后cost反而更低,但實(shí)際時(shí)間竟然長了這么多。而當(dāng)我們將日志中獲取的綁定變量值帶入SQL中再去查看執(zhí)行計(jì)劃時(shí),仍然是走索引掃描。既然如此,那比較容易想到的就是plan cache導(dǎo)致的執(zhí)行計(jì)劃錯(cuò)誤了。
由于在PostgreSQL中執(zhí)行計(jì)劃緩存只是會(huì)話級別的,PostgreSQL在生成執(zhí)行計(jì)劃緩存前,會(huì)先走5次custom plan,然后記錄這5次總的custom plan的cost, 以及custom plan的次數(shù),最后生成通用的generic plan。
以后,每次bind時(shí),會(huì)根據(jù)緩存的執(zhí)行計(jì)劃以及給定的參數(shù)值計(jì)算一個(gè)COST,如果這個(gè)COST 小于前面存儲(chǔ)的custom plan cost的平均值,則使用當(dāng)前緩存的執(zhí)行計(jì)劃。如果這個(gè)COST大于前面存儲(chǔ)的custom plan cost的平均值,則使用custom plan(即重新生成執(zhí)行計(jì)劃),同時(shí)custom plan的次數(shù)加1,custom plan總成本也會(huì)累加進(jìn)去。
既然如此,我們使用prepare語句再測試一次:
bill=# prepare p1 as select * from t1 where id = 999
bill-# and (case $1 when 'true' then info = $2 else info = $3 end) limit 1;
PREPARE
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.831..0.831 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.830..0.830 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.971 ms
Execution Time: 0.889 ms
(6 rows)
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.038..0.039 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.037..0.037 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.240 ms
Execution Time: 0.088 ms
(6 rows)
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.036..0.036 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.035..0.035 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.136 ms
Execution Time: 0.076 ms
(6 rows)
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.051..0.051 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.049..0.050 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.165 ms
Execution Time: 0.091 ms
(6 rows)
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..3.18 rows=1 width=45) (actual time=0.027..0.027 rows=0 loops=1)
-> Index Scan using idx_t1 on t1 (cost=0.56..3.18 rows=1 width=45) (actual time=0.025..0.026 rows=0 loops=1)
Index Cond: (info = 'bill'::text)
Filter: (id = 999)
Planning Time: 0.158 ms
Execution Time: 0.057 ms
(6 rows)
bill=# explain analyze execute p1('true','bill','postgres');
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
Limit (cost=0.00..0.35 rows=1 width=45) (actual time=487.564..487.564 rows=0 loops=1)
-> Seq Scan on t1 (cost=0.00..170895.98 rows=491791 width=45) (actual time=487.562..487.562 rows=0 loops=1)
Filter: ((id = 999) AND CASE $1 WHEN 'true'::text THEN (info = $2) ELSE (info = $3) END)
Rows Removed by Filter: 6000000
Planning Time: 0.119 ms
Execution Time: 487.595 ms
(6 rows)
果然在第6次時(shí)出現(xiàn)了我們想要的結(jié)果!
可以看到前5次索引掃描的cost都是3.18,而全表掃描的cost卻是0.35,所以自然優(yōu)化器選擇了全表掃描,可為什么cost變低了反而時(shí)間更久了呢?解答這個(gè)問題前我們先要來了解下limit子句的cost是如何計(jì)算的。
limit cost計(jì)算方法:
先從一個(gè)最簡單的例子看起:
我們只取1條記錄,cost很低,時(shí)間也很少。
bill=# explain analyze select * from t1 limit 1;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------
Limit (cost=0.00..0.02 rows=1 width=45) (actual time=0.105..0.106 rows=1 loops=1)
-> Seq Scan on t1 (cost=0.00..110921.49 rows=5997449 width=45) (actual time=0.103..0.103 rows=1 loops=1)
Planning Time: 0.117 ms
Execution Time: 0.133 ms
(4 rows)
加上where條件試試呢?
cost一下子變成3703.39了,似乎也很好理解,因?yàn)槲覀冊谶M(jìn)行l(wèi)imit前要使用where條件進(jìn)行一次數(shù)據(jù)過濾,所以cost變得很高了。
bill=# explain analyze select * from t1 where id = 1000 limit 1;
QUERY PLAN
---------------------------------------------------------------------------------------------------------
Limit (cost=0.00..3703.39 rows=1 width=45) (actual time=0.482..0.483 rows=1 loops=1)
-> Seq Scan on t1 (cost=0.00..125915.11 rows=34 width=45) (actual time=0.480..0.481 rows=1 loops=1)
Filter: (id = 1000)
Rows Removed by Filter: 1008
Planning Time: 0.117 ms
Execution Time: 0.523 ms
(6 rows)
但當(dāng)我們換個(gè)條件時(shí)結(jié)果又不同了:
從where id=1000變成 id=999,cost竟然一下子又降低到0.13了,似乎找到了前面全表掃描的limit cost比索引掃描還低的原因了。
bill=# explain analyze select * from t1 where id = 999 limit 1;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------
Limit (cost=0.00..0.13 rows=1 width=45) (actual time=0.041..0.042 rows=1 loops=1)
-> Seq Scan on t1 (cost=0.00..125915.11 rows=983582 width=45) (actual time=0.040..0.040 rows=1 loops=1)
Filter: (id = 999)
Rows Removed by Filter: 107
Planning Time: 0.114 ms
Execution Time: 0.079 ms
(6 rows)
那么這個(gè)limit的cost究竟是如何計(jì)算的呢,為什么條件不同cost能差這么多呢?
下面給出limit cost計(jì)算方法:
limit_cost = ( N / B ) * A
N:表示limit取的數(shù)據(jù),如limit 1則N=1;
B:表示估算得到的總記錄數(shù);
A:表示估算的總成本。
例如上面cost=0.13的執(zhí)行計(jì)劃中,N = 1,B = 983582,A = 125915.11,那么limit cost便是:
(1/983582)*125915.11 = 0.128,即執(zhí)行計(jì)劃中顯示的0.13。
簡而言之就是如果通過where條件篩選得到的行數(shù)越多,那么limit cost就會(huì)越低。
知道了這些我們再回過頭去看那條SQL就清楚了,因?yàn)閣here id = 999這個(gè)條件的數(shù)據(jù)比較多,這也就導(dǎo)致了即使是全表掃描limit cost也很低,甚至比索引掃描還低。
SELECT *
FROM t1
WHERE id = 999
AND (case $1
WHEN 'true' THEN
info = $2
ELSE info = $3 end) limit 1;
但是需要注意的是,我們即使使用explain analyze看到的執(zhí)行計(jì)劃中的cost也是一個(gè)估算值,并不是實(shí)際值,盡管這個(gè)和實(shí)際值差距不會(huì)很大,但如果cost本身就很小,那么還是會(huì)帶來一點(diǎn)誤解的。
例如前面的SQL我想要提高全表掃描的limit cost讓其大于索引掃描,這樣優(yōu)化器便會(huì)一直選擇索引掃描了,于是我將limit 1改成limit 100(即增加N的值),但是卻仍然沒有起作用:
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Limit (cost=0.56..5.58 rows=1 width=53) (actual time=0.049..0.051 rows=1 loops=1)
-> Index Scan using idx_scm_bind_scm_customer_id_index on scm_bind t (cost=0.56..5.58 rows=1 width=53) (actual time=0.049..0.050 rows=1 loops=1)
Index Cond: ((scm_customer_id)::text = 'wmGAgeDQAAXcpcw9QWkDOUQsIDI1xOqQ'::text)
Filter: ((bind_status)::text = '2'::text)
Planning Time: 0.160 ms
Execution Time: 0.072 ms
(6 rows)
Time: 0.470 ms
QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Limit (cost=0.00..8.90 rows=100 width=53) (actual time=1047.859..16654.360 rows=1 loops=1)
-> Seq Scan on scm_bind t (cost=0.00..552392.00 rows=6208050 width=53) (actual time=1047.858..16654.357 rows=1 loops=1)
Filter: (((bind_status)::text = '2'::text) AND CASE $1 WHEN 'client'::text THEN ((scm_customer_id)::text = ($2)::text) ELSE ((scm_customer_id)::text = ($3)::text) END)
Rows Removed by Filter: 12169268
Planning Time: 0.147 ms
Execution Time: 16654.459 ms
(6 rows)
Time: 16654.924 ms (00:16.655)
下面的全表掃描是第6次傳入?yún)?shù)得到的,可以看到全表掃描的cost是8.9,而索引掃描是5.58,那應(yīng)該不會(huì)選擇cost更高的8.9啊?
而當(dāng)我們?nèi)ジ檶?shí)際的cost就可以發(fā)現(xiàn):
$1 = {magic = 195726186, raw_parse_tree = 0x15df470,
query_string = 0x16d65b8 "PREPARE p1(varchar,varchar,varchar) as\n select\n t.scm_sale_customer_id,\n t.scm_customer_id\n from\n scm_bind t\n where t.bind_status = '2'\n and (case $1 when 'client' then scm_customer_id ="..., commandTag = 0x95b5ba "SELECT", param_types = 0x16d66c8, num_params = 3, parserSetup = 0x0, parserSetupArg = 0x0, cursor_options = 256, fixed_result = true,
resultDesc = 0x16d66e8, context = 0x15df250, query_list = 0x16dbe80, relationOids = 0x16e6138, invalItems = 0x0, search_path = 0x16e6168, query_context = 0x16dbd70, rewriteRoleId = 10,
rewriteRowSecurity = true, dependsOnRLS = false, gplan = 0x16ff668, is_oneshot = false, is_complete = true, is_saved = true, is_valid = true, generation = 6, next_saved = 0x0,
generic_cost = 8.8979953447539888, total_custom_cost = 52.899999999999999, num_custom_plans = 5}
實(shí)際索引掃描的cost大約數(shù)10.58,和執(zhí)行計(jì)劃中顯示的還是有一定差距的。
讓我們言歸正傳,既然知道了為什么全表掃描的limit cost更低,我們再來解決下一個(gè)問題:為什么cost很低但實(shí)際執(zhí)行時(shí)間卻這么長?
讓我們再看看執(zhí)行計(jì)劃:
Limit (cost=0.00..0.35 rows=1 width=45) (actual time=487.564..487.564 rows=0 loops=1)
-> Seq Scan on t1 (cost=0.00..170895.98 rows=491791 width=45) (actual time=487.562..487.562 rows=0 loops=1)
Filter: ((id = 999) AND CASE $1 WHEN 'true'::text THEN (info = $2) ELSE (info = $3) END)
Rows Removed by Filter: 6000000
Planning Time: 0.119 ms
Execution Time: 487.595 ms
(6 rows)
仔細(xì)觀察可以發(fā)現(xiàn),原先應(yīng)該作為索引的info列的過濾條件,竟然整個(gè)作為了filter條件去進(jìn)行數(shù)據(jù)過濾了。
那么最后的問題就出現(xiàn)在這個(gè)where條件中的case when表達(dá)式了,因?yàn)樵赾ase when表達(dá)式進(jìn)行過濾前,綁定變量還沒有傳入實(shí)際的值,而優(yōu)化器對于不確定的值自然無法選擇是否去走索引了,這里不得不吐槽一下這種寫法。。。
因此對于優(yōu)化器計(jì)算limit cost時(shí),只知道where id = 999會(huì)得到大量的數(shù)據(jù),而無法判斷后面的case when里面會(huì)得到多少數(shù)據(jù),因此雖然后面的條件只會(huì)得到很少一部分?jǐn)?shù)據(jù),但是優(yōu)化器生成limit cost時(shí)估算得到的總記錄數(shù)B只是根據(jù)id = 999去判斷,導(dǎo)致估算的cost很低,但實(shí)際卻只得到很少的數(shù)據(jù),要去表中過濾大量數(shù)據(jù)。
不得不感嘆這個(gè)“簡單”的SQL竟然包含著這么多知識(shí)。
原文鏈接:https://foucus.blog.csdn.net/article/details/122107557
相關(guān)推薦
- 2022-05-08 Python集合的增刪改查操作_python
- 2022-02-17 如何通過一道題,全方位地考察自己,是否已經(jīng)完美掌握了:this指向、作用域&作用域鏈、閉包、
- 2022-06-25 在Nginx服務(wù)器上安裝SSL證書完成HTTPS請求的步驟詳解(springboot項(xiàng)目)_ngin
- 2022-08-02 C#?HttpClient?Post參數(shù)同時(shí)上傳文件的實(shí)現(xiàn)_C#教程
- 2022-09-30 關(guān)于react中useCallback的用法_React
- 2022-07-03 C#入門之定義類成員與接口實(shí)現(xiàn)_C#教程
- 2022-11-27 C語言中花式退出程序的方式總結(jié)_C 語言
- 2022-05-05 C#?模式匹配完全指南_C#教程
- 最近更新
-
- window11 系統(tǒng)安裝 yarn
- 超詳細(xì)win安裝深度學(xué)習(xí)環(huán)境2025年最新版(
- Linux 中運(yùn)行的top命令 怎么退出?
- MySQL 中decimal 的用法? 存儲(chǔ)小
- get 、set 、toString 方法的使
- @Resource和 @Autowired注解
- Java基礎(chǔ)操作-- 運(yùn)算符,流程控制 Flo
- 1. Int 和Integer 的區(qū)別,Jav
- spring @retryable不生效的一種
- Spring Security之認(rèn)證信息的處理
- Spring Security之認(rèn)證過濾器
- Spring Security概述快速入門
- Spring Security之配置體系
- 【SpringBoot】SpringCache
- Spring Security之基于方法配置權(quán)
- redisson分布式鎖中waittime的設(shè)
- maven:解決release錯(cuò)誤:Artif
- restTemplate使用總結(jié)
- Spring Security之安全異常處理
- MybatisPlus優(yōu)雅實(shí)現(xiàn)加密?
- Spring ioc容器與Bean的生命周期。
- 【探索SpringCloud】服務(wù)發(fā)現(xiàn)-Nac
- Spring Security之基于HttpR
- Redis 底層數(shù)據(jù)結(jié)構(gòu)-簡單動(dòng)態(tài)字符串(SD
- arthas操作spring被代理目標(biāo)對象命令
- Spring中的單例模式應(yīng)用詳解
- 聊聊消息隊(duì)列,發(fā)送消息的4種方式
- bootspring第三方資源配置管理
- GIT同步修改后的遠(yuǎn)程分支