Answers:
我确实需要知道这一点,因此我对这两种方法进行了基准测试。我通常发现IN
比使用快得多OR
。
不要相信给出意见的人,科学就是测试和证据。
我运行了一个1000倍等效查询的循环(出于一致性考虑,我使用sql_no_cache
):
IN
:2.34969592094s
OR
:5.83781504631s
更新:(
我没有原始测试的源代码,就像6年前一样,尽管它返回的结果与此测试范围相同)
在要求一些示例代码进行测试时,这是最简单的用例。使用Eloquent简化语法,等效的原始SQL执行相同的操作。
$t = microtime(true);
for($i=0; $i<10000; $i++):
$q = DB::table('users')->where('id',1)
->orWhere('id',2)
->orWhere('id',3)
->orWhere('id',4)
->orWhere('id',5)
->orWhere('id',6)
->orWhere('id',7)
->orWhere('id',8)
->orWhere('id',9)
->orWhere('id',10)
->orWhere('id',11)
->orWhere('id',12)
->orWhere('id',13)
->orWhere('id',14)
->orWhere('id',15)
->orWhere('id',16)
->orWhere('id',17)
->orWhere('id',18)
->orWhere('id',19)
->orWhere('id',20)->get();
endfor;
$t2 = microtime(true);
echo $t."\n".$t2."\n".($t2-$t)."\n";
1482080514.3635
1482080517.3713
3.0078368186951
$t = microtime(true);
for($i=0; $i<10000; $i++):
$q = DB::table('users')->whereIn('id',[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20])->get();
endfor;
$t2 = microtime(true);
echo $t."\n".$t2."\n".($t2-$t)."\n";
1482080534.0185
1482080536.178
2.1595389842987
IN
语句比的速度快30%OR
。
Do not believe people who give their "opinion"
您是100%正确的,不幸的是,堆栈溢满了
Returns 1 if expr is equal to any of the values in the IN list, else returns 0. If all values are constants, they are evaluated according to the type of expr and sorted. The search for the item then is done using a binary search. This means
IN is very quick if the IN value list consists entirely of constants
. Otherwise, type conversion takes place according to the rules described at Type Conversion, but applied to all the arguments.
IN
我还对未来的Google员工进行了测试。返回结果总数为7264,共10000
SELECT * FROM item WHERE id = 1 OR id = 2 ... id = 10000
这个查询花了0.1239
几秒钟
SELECT * FROM item WHERE id IN (1,2,3,...10000)
这个查询花了0.0433
几秒钟
IN
比3倍快 OR
OR
:最好使用尽可能紧凑的表达式来表达它会产生很多多余的文本。
接受的答案没有说明原因。
下面引用了高性能MySQL第三版。
在许多数据库服务器中,IN()只是多个OR子句的同义词,因为两者在逻辑上是等效的。在MySQL中不是这样,MySQL对IN()列表中的值进行排序,并使用快速的二进制搜索来查看列表中是否包含值。列表的大小为O(Log n),而等效的OR子句序列的列表的大小为O(n)(即,大列表的速度要慢得多)
我认为BETWEEN将更快,因为应该将其转换为:
Field >= 0 AND Field <= 5
据我了解,IN仍将转换为一堆OR语句。IN的值是易于使用。(省去了必须多次键入每个列名的麻烦,并且还使它更易于与现有逻辑一起使用-您不必担心AND / OR优先级,因为IN是一个语句。使用一堆OR语句,您可以以确保用括号括起来,以确保它们被视为一种条件。)
您问题的唯一真正答案是“ 查询个人资料”。然后,您将知道哪种方法最适合您的特定情况。
这取决于你在做什么。范围是多少,数据类型是什么(我知道您的示例使用数字数据类型,但是您的问题也可以适用于许多不同的数据类型)。
这是一个您想同时编写查询的实例。使它工作,然后使用EXPLAIN找出执行差异。
我确定对此有一个具体的答案,但是实际上,这就是我如何为给定问题找出答案。
这可能会有所帮助:http : //forge.mysql.com/wiki/Top10SQLPerformanceTips
问候,
弗兰克
就在您认为安全的时候...
您的价值是eq_range_index_dive_limit
什么?特别是,该IN
子句中有更多还是更少的项目?
这将不包括基准测试,但会深入了解内部工作原理。让我们使用一个工具查看正在发生的事情-Optimizer Trace。
查询: SELECT * FROM canada WHERE id ...
使用OR
3个值中的一个,部分跟踪如下所示:
"condition_processing": {
"condition": "WHERE",
"original_condition": "((`canada`.`id` = 296172) or (`canada`.`id` = 295093) or (`canada`.`id` = 293626))",
"steps": [
{
"transformation": "equality_propagation",
"resulting_condition": "(multiple equal(296172, `canada`.`id`) or multiple equal(295093, `canada`.`id`) or multiple equal(293626, `canada`.`id`))"
},
...
"analyzing_range_alternatives": {
"range_scan_alternatives": [
{
"index": "id",
"ranges": [
"293626 <= id <= 293626",
"295093 <= id <= 295093",
"296172 <= id <= 296172"
],
"index_dives_for_eq_ranges": true,
"chosen": true
...
"refine_plan": [
{
"table": "`canada`",
"pushed_index_condition": "((`canada`.`id` = 296172) or (`canada`.`id` = 295093) or (`canada`.`id` = 293626))",
"table_condition_attached": null,
"access_type": "range"
}
]
请注意ICP是如何给出的ORs
。这意味着这OR
是不是变成IN
,和InnoDB将表演一堆=
通过ICP测试。(我认为不值得考虑使用MyISAM。)
(这是Percona的5 id
.6. 22-71.0日志;是辅助索引。)
现在是IN()的一些值
eq_range_index_dive_limit
= 10; 有8个值。
"condition_processing": {
"condition": "WHERE",
"original_condition": "(`canada`.`id` in (296172,295093,293626,295573,297148,296127,295588,295810))",
"steps": [
{
"transformation": "equality_propagation",
"resulting_condition": "(`canada`.`id` in (296172,295093,293626,295573,297148,296127,295588,295810))"
},
...
"analyzing_range_alternatives": {
"range_scan_alternatives": [
{
"index": "id",
"ranges": [
"293626 <= id <= 293626",
"295093 <= id <= 295093",
"295573 <= id <= 295573",
"295588 <= id <= 295588",
"295810 <= id <= 295810",
"296127 <= id <= 296127",
"296172 <= id <= 296172",
"297148 <= id <= 297148"
],
"index_dives_for_eq_ranges": true,
"chosen": true
...
"refine_plan": [
{
"table": "`canada`",
"pushed_index_condition": "(`canada`.`id` in (296172,295093,293626,295573,297148,296127,295588,295810))",
"table_condition_attached": null,
"access_type": "range"
}
]
请注意,IN
似乎并没有变成OR
。
注意:请注意,常数值已排序。这有两个方面的好处:
最后,IN()具有很多值
{
"condition_processing": {
"condition": "WHERE",
"original_condition": "(`canada`.`id` in (293831,292259,292881,293440,292558,295792,292293,292593,294337,295430,295034,297060,293811,295587,294651,295559,293213,295742,292605,296018,294529,296711,293919,294732,294689,295540,293000,296916,294433,297112,293815,292522,296816,293320,293232,295369,291894,293700,291839,293049,292738,294895,294473,294023,294173,293019,291976,294923,294797,296958,294075,293450,296952,297185,295351,295736,296312,294330,292717,294638,294713,297176,295896,295137,296573,292236,294966,296642,296073,295903,293057,294628,292639,293803,294470,295353,297196,291752,296118,296964,296185,295338,295956,296064,295039,297201,297136,295206,295986,292172,294803,294480,294706,296975,296604,294493,293181,292526,293354,292374,292344,293744,294165,295082,296203,291918,295211,294289,294877,293120,295387))",
"steps": [
{
"transformation": "equality_propagation",
"resulting_condition": "(`canada`.`id` in (293831,292259,292881,293440,292558,295792,292293,292593,294337,295430,295034,297060,293811,295587,294651,295559,293213,295742,292605,296018,294529,296711,293919,294732,294689,295540,293000,296916,294433,297112,293815,292522,296816,293320,293232,295369,291894,293700,291839,293049,292738,294895,294473,294023,294173,293019,291976,294923,294797,296958,294075,293450,296952,297185,295351,295736,296312,294330,292717,294638,294713,297176,295896,295137,296573,292236,294966,296642,296073,295903,293057,294628,292639,293803,294470,295353,297196,291752,296118,296964,296185,295338,295956,296064,295039,297201,297136,295206,295986,292172,294803,294480,294706,296975,296604,294493,293181,292526,293354,292374,292344,293744,294165,295082,296203,291918,295211,294289,294877,293120,295387))"
},
...
"analyzing_range_alternatives": {
"range_scan_alternatives": [
{
"index": "id",
"ranges": [
"291752 <= id <= 291752",
"291839 <= id <= 291839",
...
"297196 <= id <= 297196",
"297201 <= id <= 297201"
],
"index_dives_for_eq_ranges": false,
"rows": 111,
"chosen": true
...
"refine_plan": [
{
"table": "`canada`",
"pushed_index_condition": "(`canada`.`id` in (293831,292259,292881,293440,292558,295792,292293,292593,294337,295430,295034,297060,293811,295587,294651,295559,293213,295742,292605,296018,294529,296711,293919,294732,294689,295540,293000,296916,294433,297112,293815,292522,296816,293320,293232,295369,291894,293700,291839,293049,292738,294895,294473,294023,294173,293019,291976,294923,294797,296958,294075,293450,296952,297185,295351,295736,296312,294330,292717,294638,294713,297176,295896,295137,296573,292236,294966,296642,296073,295903,293057,294628,292639,293803,294470,295353,297196,291752,296118,296964,296185,295338,295956,296064,295039,297201,297136,295206,295986,292172,294803,294480,294706,296975,296604,294493,293181,292526,293354,292374,292344,293744,294165,295082,296203,291918,295211,294289,294877,293120,295387))",
"table_condition_attached": null,
"access_type": "range"
}
]
旁注:由于跟踪的庞大性,我需要这样做:
@@global.optimizer_trace_max_mem_size = 32222;
以下是使用MySQL 5.6 @SQLFiddle的6个查询的详细信息
总之,这6个查询涵盖了独立索引的列,每种数据类型使用了2个查询。无论使用IN()还是OR,所有查询都会导致使用索引。
| ORs | IN()
integer | uses index | uses index
date | uses index | uses index
varchar | uses index | uses index
我真的只是想揭穿OR表示不能使用索引的声明。这不是真的 可以在使用OR的查询中使用索引,如以下示例中显示的6个查询所示。
在我看来,很多人也忽略了IN()是一组OR的语法快捷方式的事实。在小规模性能上,使用IN()-v- OR之间的差异非常大(无穷)。
虽然在更大的范围内IN()当然更方便,但是从逻辑上讲它等于一组OR条件。每个查询的情况都会发生变化,因此始终最好在表上测试查询。
6个解释计划的摘要,所有“使用索引条件”(向右滚动)
Query select_type table type possible_keys key key_len ref rows filtered Extra
------------- --------- ------- --------------- ----------- --------- ----- ------ ---------- -----------------------
Integers using OR SIMPLE mytable range aNum_idx aNum_idx 4 10 100.00 Using index condition
Integers using IN SIMPLE mytable range aNum_idx aNum_idx 4 10 100.00 Using index condition
Dates using OR SIMPLE mytable range aDate_idx aDate_idx 6 7 100.00 Using index condition
Dates using IN SIMPLE mytable range aDate_idx aDate_idx 6 7 100.00 Using index condition
Varchar using OR SIMPLE mytable range aName_idx aName_idx 768 10 100.00 Using index condition
Varchar using IN SIMPLE mytable range aName_idx aName_idx 768 10 100.00 Using index condition
MySQL 5.6模式设置:
CREATE TABLE `myTable` (
`id` mediumint(8) unsigned NOT NULL auto_increment,
`aName` varchar(255) default NULL,
`aDate` datetime,
`aNum` mediumint(8),
PRIMARY KEY (`id`)
) AUTO_INCREMENT=1;
ALTER TABLE `myTable` ADD INDEX `aName_idx` (`aName`);
ALTER TABLE `myTable` ADD INDEX `aDate_idx` (`aDate`);
ALTER TABLE `myTable` ADD INDEX `aNum_idx` (`aNum`);
INSERT INTO `myTable` (`aName`,`aDate`)
VALUES
("Daniel","2017-09-19 01:22:31")
,("Quentin","2017-06-03 01:06:45")
,("Chester","2017-06-14 17:49:36")
,("Lev","2017-08-30 06:27:59")
,("Garrett","2018-10-04 02:40:37")
,("Lane","2017-01-22 17:11:21")
,("Chaim","2017-09-20 11:13:46")
,("Kieran","2018-03-10 18:37:26")
,("Cedric","2017-05-20 16:25:10")
,("Conan","2018-07-10 06:29:39")
,("Rudyard","2017-07-14 00:04:00")
,("Chadwick","2018-08-18 08:54:08")
,("Darius","2018-10-02 06:55:56")
,("Joseph","2017-06-19 13:20:33")
,("Wayne","2017-04-02 23:20:25")
,("Hall","2017-10-13 00:17:24")
,("Craig","2016-12-04 08:15:22")
,("Keane","2018-03-12 04:21:46")
,("Russell","2017-07-14 17:21:58")
,("Seth","2018-07-25 05:51:30")
,("Cole","2018-06-09 15:32:53")
,("Donovan","2017-08-12 05:21:35")
,("Damon","2017-06-27 03:44:19")
,("Brian","2017-02-01 23:35:20")
,("Harper","2017-08-25 04:29:27")
,("Chandler","2017-09-30 23:54:06")
,("Edward","2018-07-30 12:18:07")
,("Curran","2018-05-23 09:31:53")
,("Uriel","2017-05-08 03:31:43")
,("Honorato","2018-04-07 14:57:53")
,("Griffin","2017-01-07 23:35:31")
,("Hasad","2017-05-15 05:32:41")
,("Burke","2017-07-04 01:11:19")
,("Hyatt","2017-03-14 17:12:28")
,("Brenden","2017-10-17 05:16:14")
,("Ryan","2018-10-10 08:07:55")
,("Giacomo","2018-10-06 14:21:21")
,("James","2018-02-06 02:45:59")
,("Colt","2017-10-10 08:11:26")
,("Kermit","2017-09-18 16:57:16")
,("Drake","2018-05-20 22:08:36")
,("Berk","2017-04-16 17:39:32")
,("Alan","2018-09-01 05:33:05")
,("Deacon","2017-04-20 07:03:05")
,("Omar","2018-03-02 15:04:32")
,("Thaddeus","2017-09-19 04:07:54")
,("Troy","2016-12-13 04:24:08")
,("Rogan","2017-11-02 00:03:25")
,("Grant","2017-08-21 01:45:16")
,("Walker","2016-11-26 15:54:52")
,("Clarke","2017-07-20 02:26:56")
,("Clayton","2018-08-16 05:09:29")
,("Denton","2018-08-11 05:26:05")
,("Nicholas","2018-07-19 09:29:55")
,("Hashim","2018-08-10 20:38:06")
,("Todd","2016-10-25 01:01:36")
,("Xenos","2017-05-11 22:50:35")
,("Bert","2017-06-17 18:08:21")
,("Oleg","2018-01-03 13:10:32")
,("Hall","2018-06-04 01:53:45")
,("Evan","2017-01-16 01:04:25")
,("Mohammad","2016-11-18 05:42:52")
,("Armand","2016-12-18 06:57:57")
,("Kaseem","2018-06-12 23:09:57")
,("Colin","2017-06-29 05:25:52")
,("Arthur","2016-12-29 04:38:13")
,("Xander","2016-11-14 19:35:32")
,("Dante","2016-12-01 09:01:04")
,("Zahir","2018-02-17 14:44:53")
,("Raymond","2017-03-09 05:33:06")
,("Giacomo","2017-04-17 06:12:52")
,("Fulton","2017-06-04 00:41:57")
,("Chase","2018-01-14 03:03:57")
,("William","2017-05-08 09:44:59")
,("Fuller","2017-03-31 20:35:20")
,("Jarrod","2017-02-15 02:45:29")
,("Nissim","2018-03-11 14:19:25")
,("Chester","2017-11-05 00:14:27")
,("Perry","2017-12-24 11:58:04")
,("Theodore","2017-06-26 12:34:12")
,("Mason","2017-10-02 03:53:49")
,("Brenden","2018-10-08 10:09:47")
,("Jerome","2017-11-05 20:34:25")
,("Keaton","2018-08-18 00:55:56")
,("Tiger","2017-05-21 16:59:07")
,("Benjamin","2018-04-10 14:46:36")
,("John","2018-09-05 18:53:03")
,("Jakeem","2018-10-11 00:17:38")
,("Kenyon","2017-12-18 22:19:29")
,("Ferris","2017-03-29 06:59:13")
,("Hoyt","2017-01-03 03:48:56")
,("Fitzgerald","2017-07-27 11:27:52")
,("Forrest","2017-10-05 23:14:21")
,("Jordan","2017-01-11 03:48:09")
,("Lev","2017-05-25 08:03:39")
,("Chase","2017-06-18 19:09:23")
,("Ryder","2016-12-13 12:50:50")
,("Malik","2017-11-19 15:15:55")
,("Zeph","2018-04-04 11:22:12")
,("Amala","2017-01-29 07:52:17")
;
。
update MyTable
set aNum = id
;
查询1:
select 'aNum by OR' q, mytable.*
from mytable
where aNum = 12
OR aNum = 22
OR aNum = 27
OR aNum = 32
OR aNum = 42
OR aNum = 52
OR aNum = 62
OR aNum = 65
OR aNum = 72
OR aNum = 82
结果:
| q | id | aName | aDate | aNum |
|------------|----|----------|----------------------|------|
| aNum by OR | 12 | Chadwick | 2018-08-18T08:54:08Z | 12 |
| aNum by OR | 22 | Donovan | 2017-08-12T05:21:35Z | 22 |
| aNum by OR | 27 | Edward | 2018-07-30T12:18:07Z | 27 |
| aNum by OR | 32 | Hasad | 2017-05-15T05:32:41Z | 32 |
| aNum by OR | 42 | Berk | 2017-04-16T17:39:32Z | 42 |
| aNum by OR | 52 | Clayton | 2018-08-16T05:09:29Z | 52 |
| aNum by OR | 62 | Mohammad | 2016-11-18T05:42:52Z | 62 |
| aNum by OR | 65 | Colin | 2017-06-29T05:25:52Z | 65 |
| aNum by OR | 72 | Fulton | 2017-06-04T00:41:57Z | 72 |
| aNum by OR | 82 | Brenden | 2018-10-08T10:09:47Z | 82 |
查询2:
select 'aNum by IN' q, mytable.*
from mytable
where aNum IN (
12
, 22
, 27
, 32
, 42
, 52
, 62
, 65
, 72
, 82
)
结果:
| q | id | aName | aDate | aNum |
|------------|----|----------|----------------------|------|
| aNum by IN | 12 | Chadwick | 2018-08-18T08:54:08Z | 12 |
| aNum by IN | 22 | Donovan | 2017-08-12T05:21:35Z | 22 |
| aNum by IN | 27 | Edward | 2018-07-30T12:18:07Z | 27 |
| aNum by IN | 32 | Hasad | 2017-05-15T05:32:41Z | 32 |
| aNum by IN | 42 | Berk | 2017-04-16T17:39:32Z | 42 |
| aNum by IN | 52 | Clayton | 2018-08-16T05:09:29Z | 52 |
| aNum by IN | 62 | Mohammad | 2016-11-18T05:42:52Z | 62 |
| aNum by IN | 65 | Colin | 2017-06-29T05:25:52Z | 65 |
| aNum by IN | 72 | Fulton | 2017-06-04T00:41:57Z | 72 |
| aNum by IN | 82 | Brenden | 2018-10-08T10:09:47Z | 82 |
查询3:
select 'adate by OR' q, mytable.*
from mytable
where aDate= str_to_date("2017-02-15 02:45:29",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2018-03-10 18:37:26",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2017-05-20 16:25:10",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2018-07-10 06:29:39",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2017-07-14 00:04:00",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2018-08-18 08:54:08",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2018-10-02 06:55:56",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2017-04-20 07:03:05",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2018-03-02 15:04:32",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2017-09-19 04:07:54",'%Y-%m-%d %h:%i:%s')
OR aDate = str_to_date("2016-12-13 04:24:08",'%Y-%m-%d %h:%i:%s')
结果:
| q | id | aName | aDate | aNum |
|-------------|----|----------|----------------------|------|
| adate by OR | 47 | Troy | 2016-12-13T04:24:08Z | 47 |
| adate by OR | 76 | Jarrod | 2017-02-15T02:45:29Z | 76 |
| adate by OR | 44 | Deacon | 2017-04-20T07:03:05Z | 44 |
| adate by OR | 46 | Thaddeus | 2017-09-19T04:07:54Z | 46 |
| adate by OR | 10 | Conan | 2018-07-10T06:29:39Z | 10 |
| adate by OR | 12 | Chadwick | 2018-08-18T08:54:08Z | 12 |
| adate by OR | 13 | Darius | 2018-10-02T06:55:56Z | 13 |
查询4:
select 'adate by IN' q, mytable.*
from mytable
where aDate IN (
str_to_date("2017-02-15 02:45:29",'%Y-%m-%d %h:%i:%s')
, str_to_date("2018-03-10 18:37:26",'%Y-%m-%d %h:%i:%s')
, str_to_date("2017-05-20 16:25:10",'%Y-%m-%d %h:%i:%s')
, str_to_date("2018-07-10 06:29:39",'%Y-%m-%d %h:%i:%s')
, str_to_date("2017-07-14 00:04:00",'%Y-%m-%d %h:%i:%s')
, str_to_date("2018-08-18 08:54:08",'%Y-%m-%d %h:%i:%s')
, str_to_date("2018-10-02 06:55:56",'%Y-%m-%d %h:%i:%s')
, str_to_date("2017-04-20 07:03:05",'%Y-%m-%d %h:%i:%s')
, str_to_date("2018-03-02 15:04:32",'%Y-%m-%d %h:%i:%s')
, str_to_date("2017-09-19 04:07:54",'%Y-%m-%d %h:%i:%s')
, str_to_date("2016-12-13 04:24:08",'%Y-%m-%d %h:%i:%s')
)
结果:
| q | id | aName | aDate | aNum |
|-------------|----|----------|----------------------|------|
| adate by IN | 47 | Troy | 2016-12-13T04:24:08Z | 47 |
| adate by IN | 76 | Jarrod | 2017-02-15T02:45:29Z | 76 |
| adate by IN | 44 | Deacon | 2017-04-20T07:03:05Z | 44 |
| adate by IN | 46 | Thaddeus | 2017-09-19T04:07:54Z | 46 |
| adate by IN | 10 | Conan | 2018-07-10T06:29:39Z | 10 |
| adate by IN | 12 | Chadwick | 2018-08-18T08:54:08Z | 12 |
| adate by IN | 13 | Darius | 2018-10-02T06:55:56Z | 13 |
查询5:
select 'name by OR' q, mytable.*
from mytable
where aname = 'Alan'
OR aname = 'Brian'
OR aname = 'Chandler'
OR aname = 'Darius'
OR aname = 'Evan'
OR aname = 'Ferris'
OR aname = 'Giacomo'
OR aname = 'Hall'
OR aname = 'James'
OR aname = 'Jarrod'
结果:
| q | id | aName | aDate | aNum |
|-------------|----|----------|----------------------|------|
| name by OR | 43 | Alan | 2018-09-01T05:33:05Z | 43 |
| name by OR | 24 | Brian | 2017-02-01T23:35:20Z | 24 |
| name by OR | 26 | Chandler | 2017-09-30T23:54:06Z | 26 |
| name by OR | 13 | Darius | 2018-10-02T06:55:56Z | 13 |
| name by OR | 61 | Evan | 2017-01-16T01:04:25Z | 61 |
| name by OR | 90 | Ferris | 2017-03-29T06:59:13Z | 90 |
| name by OR | 37 | Giacomo | 2018-10-06T14:21:21Z | 37 |
| name by OR | 71 | Giacomo | 2017-04-17T06:12:52Z | 71 |
| name by OR | 16 | Hall | 2017-10-13T00:17:24Z | 16 |
| name by OR | 60 | Hall | 2018-06-04T01:53:45Z | 60 |
| name by OR | 38 | James | 2018-02-06T02:45:59Z | 38 |
| name by OR | 76 | Jarrod | 2017-02-15T02:45:29Z | 76 |
查询6:
select 'name by IN' q, mytable.*
from mytable
where aname IN (
'Alan'
,'Brian'
,'Chandler'
, 'Darius'
, 'Evan'
, 'Ferris'
, 'Giacomo'
, 'Hall'
, 'James'
, 'Jarrod'
)
结果:
| q | id | aName | aDate | aNum |
|------------|----|----------|----------------------|------|
| name by IN | 43 | Alan | 2018-09-01T05:33:05Z | 43 |
| name by IN | 24 | Brian | 2017-02-01T23:35:20Z | 24 |
| name by IN | 26 | Chandler | 2017-09-30T23:54:06Z | 26 |
| name by IN | 13 | Darius | 2018-10-02T06:55:56Z | 13 |
| name by IN | 61 | Evan | 2017-01-16T01:04:25Z | 61 |
| name by IN | 90 | Ferris | 2017-03-29T06:59:13Z | 90 |
| name by IN | 37 | Giacomo | 2018-10-06T14:21:21Z | 37 |
| name by IN | 71 | Giacomo | 2017-04-17T06:12:52Z | 71 |
| name by IN | 16 | Hall | 2017-10-13T00:17:24Z | 16 |
| name by IN | 60 | Hall | 2018-06-04T01:53:45Z | 60 |
| name by IN | 38 | James | 2018-02-06T02:45:59Z | 38 |
| name by IN | 76 | Jarrod | 2017-02-15T02:45:29Z | 76 |
我敢打赌它们是相同的,您可以通过执行以下操作来运行测试:
循环“ in(1,2,3,4)” 500次,看看需要多长时间。在“ = 1或= 2或= 3 ...”版本上循环500次,并查看其运行时间。
您还可以尝试一种联接方式,如果someField是索引并且您的表很大,则可能会更快...
SELECT ...
FROM ...
INNER JOIN (SELECT 1 as newField UNION ALL SELECT 2 UNION ALL SELECT 3 UNION ALL SELECT 4) dt ON someFIELD =newField
我在SQL Server上尝试了上面的join方法,它与in(1,2,3,4)几乎相同,它们都导致聚集索引查找。我不确定MySQL将如何处理它们。
s I could say that it can also be converted to UNION
的行的某些说法,建议将其替换为OR 来优化查询。