云原生数据库 PolarDB：CREATE TABLE（AUTO模式）

分区表

CREATE TABLE key_tbl(
 id bigint not null auto_increment,
 bid int,
 name varchar(30),
 birthday datetime not null,
 primary key(id)
)
PARTITION BY KEY(name, id)
PARTITIONS 8;

##（命中分区裁剪，只需扫描一个分区）
SELECT id from key_tbl where name='Jack';

CREATE TABLE hash_tbl(
 id bigint not null auto_increment,
 bid int,
 name varchar(30),
 birthday datetime not null,
 primary key(id)
)
partition by hash(id)
partitions 8;

CREATE TABLE hash_tbl_todays(
 id bigint not null auto_increment,
 bid int,
 name varchar(30),
 birthday datetime not null,
 primary key(id)
)
PARTITION BY HASH(TO_DAYS(birthday))
PARTITIONS 8;

CREATE TABLE hash_tbl2(
 id bigint not null auto_increment,
 bid int,
 name varchar(30),
 birthday datetime not null,
 primary key(id)
)
PARTITION BY HASH(name, birthday)
PARTITIONS 8;

##SQL1（命中分区裁剪，只扫描一个分区）:
SELECT id from hash_tbl2 where name='Jack' and birthday='1990-11-11';

##SQL2（没命中分区裁剪，全分区扫描）:
SELECT id from hash_tbl2 where name='Jack';

CREATE TABLE orders(
 order_id int,
 order_time datetime not null)
PARTITION BY range columns(order_id,order_time)
(
  PARTITION p1 VALUES LESS THAN (10000,'2021-01-01'),
  PARTITION p2 VALUES LESS THAN (20000,'2021-01-01'),
  PARTITION p3 VALUES LESS THAN (30000,'2021-01-01'),
  PARTITION p4 VALUES LESS THAN (40000,'2021-01-01'),
  PARTITION p5 VALUES LESS THAN (50000,'2021-01-01'),
  PARTITION p6 VALUES LESS THAN (MAXVALUE,MAXVALUE)
);

CREATE TABLE orders_todays(
 id int,
 order_time datetime not null)
PARTITION BY RANGE(to_days(order_time))
(
  PARTITION p1 VALUES LESS THAN (to_days('2021-01-01')),
  PARTITION p2 VALUES LESS THAN (to_days('2021-04-01')),
  PARTITION p3 VALUES LESS THAN (to_days('2021-07-01')),
  PARTITION p4 VALUES LESS THAN (to_days('2021-10-01')),
  PARTITION p5 VALUES LESS THAN (to_days('2022-01-01')),
  PARTITION p6 VALUES LESS THAN (MAXVALUE)
);

CREATE TABLE orders_region(
 id int,
 country varchar(64),
 city varchar(64),
 order_time datetime not null)
PARTITION BY LIST COLUMNS(country,city)
(
  PARTITION p1 VALUES IN (('China','Hangzhou'), ('China','Beijing')),
  PARTITION p2 VALUES IN (('United States','NewYork'),('United States','Chicago')),
  PARTITION p3 VALUES IN (('Russian','Moscow'))
);

CREATE TABLE orders_years(
 id int,
 country varchar(64),
 city varchar(64),
 order_time datetime not null)
PARTITION BY LIST(YEAR(order_time))
(
  PARTITION p1 VALUES IN (1990,1991,1992,1993,1994,1995,1996,1997,1998,1999),
  PARTITION p2 VALUES IN (2000,2001,2002,2003,2004,2005,2006,2007,2008,2009),
  PARTITION p3 VALUES IN (2010,2011,2012,2013,2014,2015,2016,2017,2018,2019)
);

CREATE TABLE orders_region(
 id int,
 country varchar(64),
 city varchar(64),
 order_time datetime not null)
PARTITION BY LIST COLUMNS(country,city)
(
  PARTITION p1 VALUES IN (('China','Hangzhou'), ('China','Beijing')),
  PARTITION p2 VALUES IN (('United States','NewYork'),('United States','Chicago')),
  PARTITION p3 VALUES IN (('Russian','Moscow')),
  PARTITION pd VALUES IN (DEFAULT)
);

CREATE TABLE orders_years(
 id int,
 country varchar(64),
 city varchar(64),
 order_time datetime not null)
PARTITION BY LIST(YEAR(order_time))
(
  PARTITION p1 VALUES IN (1990,1991,1992,1993,1994,1995,1996,1997,1998,1999),
  PARTITION p2 VALUES IN (2000,2001,2002,2003,2004,2005,2006,2007,2008,2009),
  PARTITION p3 VALUES IN (2010,2011,2012,2013,2014,2015,2016,2017,2018,2019),
  PARTITION pd VALUES IN (DEFAULT)
);

CREATE TABLE t_orders(
 id bigint not null auto_increment, 
 order_id bigint, 
 buyer_id bigint,
 order_time datetime not null,
 primary key(id)
) 
PARTITION BY CO_HASH(
 	RIGHT(`order_id`,6) /*取c1列的后6位字符*/,
 	RIGHT(`buyer_id`,6) /*取c2列的后6位字符*/
) 
PARTITIONS 8;

CREATE TABLE orders(
 id bigint not null auto_increment, 
 buyer_id bigint,
 order_id bigint,
 ...
 primary key(id)
) 
PARTITION BY RANGE_HASH(order_id, buyer_Id,6) 
PARTITIONS 8;

CREATE TABLE orders(
 id bigint not null auto_increment, 
 buyer_id bigint,
 order_id bigint,
 ...
 primary key(id)
) 
PARTITION BY CO_HASH(
 	RIGHT(`order_id`,6) /*取c1列的后6位字符*/,
 	RIGHT(`buyer_id`,6) /*取c2列的后6位字符*/
) 
PARTITIONS 8;

/*
 * 定义LIST-KEY的模板化子分区，
 * 一级分区按 LIST COLUMNS 分为3个分区，
 * 每个一级分区继续按 KEY 策略分为4个二级分区，
 * 因此，总共会有 12个 物理分区
*/
CREATE TABLE sp_tbl_list_key_tp(
 id int, 
 country varchar(64),
 city varchar(64),
 order_time datetime not null,
 PRIMARY KEY(id)
)
PARTITION BY LIST COLUMNS(country,city) 
SUBPARTITION BY KEY(id) SUBPARTITIONS 4
(
  PARTITION p1 VALUES IN (('China','Hangzhou')),
  PARTITION p2 VALUES IN (('Russian','Moscow')),
  PARTITION pd VALUES IN (DEFAULT)
);

/*
 * 定义LIST-KEY的非模板化子分区，
 * 一级分区按 LIST COLUMNS 分为3个分区，
 * 每个一级分区继续按 KEY 策略进行二级分区，
 * 各一级分区的二级分区数目分别定义为 2，3，4，
 * 因此，总共会有 9 个 物理分区
*/
CREATE TABLE sp_tbl_list_key_ntp(
 id int, 
 country varchar(64),
 city varchar(64),
 order_time datetime not null,
 PRIMARY KEY(id)
)
PARTITION BY LIST COLUMNS(country,city) 
SUBPARTITION BY KEY(id)
(
  PARTITION p1 VALUES IN (('China','Hangzhou')) SUBPARTITIONS 2,
  PARTITION p2 VALUES IN (('Russian','Moscow')) SUBPARTITIONS 3,
  PARTITION pd VALUES IN (DEFAULT) SUBPARTITIONS 4
);

默认自动分区

• 建表SQL在不指定分区键的情况下，PolarDB-X默认会按主键（如果表没有指定主键，则使用隐式主键）并使用KEY分区进行默认分区，自动分区创建的均为一级分区表。

  默认分区的分区数目=实例创建时逻辑节点数×8。例如，PolarDB-X实例创建时，逻辑节点是2，默认分区数目就是16。

• 除了主表默认会按主键自动分区，主表中所有索引也会默认以索引列与主键列作为分区键并进行自动分区。

如下示例是标准的MySQL建表语法，主键是id，索引列为name：

CREATE TABLE auto_part_tbl(
 id bigint not null auto_increment,
 bid int,
 name varchar(30),
 primary key(id),
 index idx_name (name)
);

若使用SHOW CREATE TABLE语句查询该建表语句，显示标准的MySQL建表语法，自动隐藏所有分区信息：

show create table auto_part_tbl;
+---------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE         | CREATE TABLE                                                                                                                                                                                                                     |
+---------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| auto_part_tbl | CREATE TABLE `auto_part_tbl` (
    `id` bigint(20) NOT NULL AUTO_INCREMENT,
    `bid` int(11) DEFAULT NULL,
    `name` varchar(30) DEFAULT NULL,
    PRIMARY KEY (`id`),
    INDEX `idx_name` (`name`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8 |
+---------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.06 sec)

若使用SHOW FULL CREATE TABLE查询该建表语句，则会显示上述主表及其索引表的所有分区信息：

show full create table auto_part_tbl;
+---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE         | CREATE TABLE                                                                                                                                                                                                                                                                                                                                                                                                              |
+---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| auto_part_tbl | CREATE PARTITION TABLE `auto_part_tbl` (
    `id` bigint(20) NOT NULL AUTO_INCREMENT,
    `bid` int(11) DEFAULT NULL,
    `name` varchar(30) DEFAULT NULL,
    PRIMARY KEY (`id`),
    GLOBAL INDEX /* idx_name_$a870 */ `idx_name` (`name`) PARTITION BY KEY (`name`, `id`) PARTITIONS 16,
    LOCAL KEY `_local_idx_name` (`name`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8
PARTITION BY KEY(`id`)
PARTITIONS 16
/* tablegroup = `tg108` */ |
+---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.03 sec)

手动分区

通过在代码中指定分区列、分区函数、分区类型等创建的表为手动分区表，详细类型请参见手动创建分区表（AUTO模式）。

数据类型说明

表 4. 各分区策略分区列数据类型的支持情况

关于分区列的数据类型与路由计算的说明

分区表的路由计算是直接依赖于分区列的数据类型的，特别是Key分区与Hash分区。因此，使用不同数据类型的分区列，其哈希值算法或比较算法（例如大小写是否敏感）的实现不一样，会产生不一样的路由行为（MySQL的分区路由算法也是类型强相关的）。

如下所示，假如tbl_int表是分区列，类型是int且分区数是1024, 而tbl_bigint分区列类型是bigint且分区数目也是1024。它们虽然都是整数类型，但是由于数据类型不一样，对于同一个查询值（12345678）的路由结果也不一样：

show create table tbl_int;
+---------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE   | CREATE TABLE                                                                                                                                                                |
+---------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| tbl_int | CREATE TABLE `tbl_int` (
    `a` int(11) NOT NULL,
    KEY `auto_shard_key_a` USING BTREE (`a`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8mb4
PARTITION BY KEY(`a`)
PARTITIONS 1024 |
+---------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.02 sec)

show create table tbl_bigint;
+------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE      | CREATE TABLE                                                                                                                                                                      |
+------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| tbl_bigint | CREATE TABLE `tbl_bigint` (
    `a` bigint(20) NOT NULL,
    KEY `auto_shard_key_a` USING BTREE (`a`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8mb4
PARTITION BY KEY(`a`)
PARTITIONS 1024 |
+------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.10 sec)mysql> create table if not exists tbl_bigint(a bigint not null)
    -> partition by key(a) partitions 1024;
Query OK, 0 rows affected (28.41 sec)

explain select * from tbl_int where a=12345678;
+---------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                             |
+---------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_int[p260]", sql="SELECT `a` FROM `tbl_int` AS `tbl_int` WHERE (`a` = ?)") |
| HitCache:false                                                                                    |
| Source:PLAN_CACHE                                                                                 |
| TemplateId: c90af636                                                                              |
+---------------------------------------------------------------------------------------------------+
4 rows in set (0.45 sec)

explain select * from tbl_bigint where a=12345678;
+------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                      |
+------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_bigint[p477]", sql="SELECT `a` FROM `tbl_bigint` AS `tbl_bigint` WHERE (`a` = ?)") |
| HitCache:false                                                                                             |
| Source:PLAN_CACHE                                                                                          |
| TemplateId: 9b2fa47c                                                                                       |
+------------------------------------------------------------------------------------------------------------+
4 rows in set (0.02 sec)
            

分区列大小写、字符集及校验集的说明

示例1

show create table tbl_varchar_cs;
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE          | CREATE TABLE                                                                                                                                                                                                          |
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| tbl_varchar_cs | CREATE TABLE `tbl_varchar_cs` (
    `a` varchar(64) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL,
    KEY `auto_shard_key_a` USING BTREE (`a`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8
PARTITION BY KEY(`a`)
PARTITIONS 64 |
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.07 sec)

explain select a from tbl_varchar_cs where a in ('AbcD');
+-------------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                                   |
+-------------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_varchar_cs[p29]", sql="SELECT `a` FROM `tbl_varchar_cs` AS `tbl_varchar_cs` WHERE (`a` IN(?))") |
| HitCache:false                                                                                                          |
| Source:PLAN_CACHE                                                                                                       |
| TemplateId: 2c49c244                                                                                                    |
+-------------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.11 sec)

explain select a from tbl_varchar_cs where a in ('abcd');
+-------------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                                   |
+-------------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_varchar_cs[p11]", sql="SELECT `a` FROM `tbl_varchar_cs` AS `tbl_varchar_cs` WHERE (`a` IN(?))") |
| HitCache:true                                                                                                           |
| Source:PLAN_CACHE                                                                                                       |
| TemplateId: 2c49c244                                                                                                    |
+-------------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.02 sec)

示例2

show create table tbl_varchar_ci;
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE          | CREATE TABLE                                                                                                                                                                      |
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| tbl_varchar_ci | CREATE TABLE `tbl_varchar_ci` (
    `a` varchar(64) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL,
    KEY `auto_shard_key_a` USING BTREE (`a`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8
PARTITION BY KEY(`a`)
PARTITIONS 64 |
+----------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.06 sec)

explain select a from tbl_varchar_ci where a in ('AbcD');
+------------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                                  |
+------------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_varchar_ci[p4]", sql="SELECT `a` FROM `tbl_varchar_ci` AS `tbl_varchar_ci` WHERE (`a` IN(?))") |
| HitCache:false                                                                                                         |
| Source:PLAN_CACHE                                                                                                      |
| TemplateId: 5c97178e                                                                                                   |
+------------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.15 sec)

explain select a from tbl_varchar_ci where a in ('abcd');
+------------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                                  |
+------------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_varchar_ci[p4]", sql="SELECT `a` FROM `tbl_varchar_ci` AS `tbl_varchar_ci` WHERE (`a` IN(?))") |
| HitCache:true                                                                                                          |
| Source:PLAN_CACHE                                                                                                      |
| TemplateId: 5c97178e                                                                                                   |
+------------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.02 sec)

分区列字符集与校验集的变更

由于分区表的路由算法与数据类型是相关的，如果分区列的字符集与校验集被修改，将会导致全表所有数据的重分布。所以请慎重地修改分区列的数据类型。

关于分区列的类型截断与类型转换的说明

分区列类型截断

SQL在查询或插入时，若指定分区列的常量表达式超过了分区列类型所能表达的范围，PolarDB-X会先产生类型截断，然后再使用类型截断后的值进行路由计算。

例如：tbl_smallint表分区列类型是smallint，smallint的正常取值范围是[-32768, 32767]。因此，如果insert的值超过smallint的范围（例如，12745678或-12345678），则会被截断类型的最大值或最小值（32767或-32768），如下所示 。

show create table tbl_smallint;
+--------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| TABLE        | CREATE TABLE                                                                                                                                                                        |
+--------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| tbl_smallint | CREATE TABLE `tbl_smallint` (
    `a` smallint(6) NOT NULL,
    KEY `auto_shard_key_a` USING BTREE (`a`)
) ENGINE = InnoDB DEFAULT CHARSET = utf8mb4
PARTITION BY KEY(`a`)
PARTITIONS 128 |
+--------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.06 sec)
set sql_mode='';
Query OK, 0 rows affected (0.00 sec)

insert into tbl_smallint values (12345678),(-12345678);
Query OK, 2 rows affected (0.07 sec)

select * from tbl_smallint;
+--------+
| a      |
+--------+
| -32768 |
|  32767 |
+--------+
2 rows in set (3.51 sec)

explain select * from tbl_smallint where a=12345678;
+------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                            |
+------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_smallint[p117]", sql="SELECT `a` FROM `tbl_smallint` AS `tbl_smallint` WHERE (`a` = ?)") |
| HitCache:false                                                                                                   |
| Source:PLAN_CACHE                                                                                                |
| TemplateId: afb464d5                                                                                             |
+------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.16 sec)

explain select * from tbl_smallint where a=32767;
+------------------------------------------------------------------------------------------------------------------+
| LOGICAL EXECUTIONPLAN                                                                                            |
+------------------------------------------------------------------------------------------------------------------+
| LogicalView(tables="tbl_smallint[p117]", sql="SELECT `a` FROM `tbl_smallint` AS `tbl_smallint` WHERE (`a` = ?)") |
| HitCache:true                                                                                                    |
| Source:PLAN_CACHE                                                                                                |
| TemplateId: afb464d5                                                                                             |
+------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.03 sec)

同理，如果查询的常量值超过类型范围，也会按截断后的值进行路由。因此，对于tbl_smallint表，a=12345678与a=32767的分区路由结果是完全相同的。

分区列类型转换

SQL在查询或插入时，若指定分区列的常量表达式与分区列的类型不一致，PolarDB-X会先对常量表达式进行隐式类型转换，然后再使用类型转换后的值进行路由计算。但是，类型转换也有可能会出现转换失败的情况，例如，字符串abc无法转换为整型。

对于分区列出现类型转换及其失败的情况，PolarDB-X按DQL、DML、DDL会有不同的行为：

• DQL（特指WHERE条件表达式中的分区列的类型转换）

  • 类型转换成功：按类型转换后的值进行分区路由；

  • 类型转换失败：该分区列条件会被直接忽略，走全表扫描。

• DML（特指Insert或Replace）

  • 类型转换成功：按类型转换后的值进行分区路由；

  • 类型转换失败：直接报错，拒绝执行。

• DDL（特指分区表相关的DDL，如建表、分裂等）

  • 类型转换成功：直接报错，拒绝执行，DDL不允许出现类型转换；

  • 类型转换失败：直接报错，拒绝执行。

与MySQL分区表的语法差异