kingshard最佳实践（二）

上篇主要介绍了kingshard的工作流程、搭建及部分功能测试，今天来测试一下kingshard对sharding的支持

 

一、kingshard的sharding介绍

1.1  kingshard的sharding优势

现在开源的MySQL Proxy已经有几款了，并且有的已经在生产环境上广泛应用。但这些proxy在sharding方面，都是不能分子表的。也就是说一个node节点只能分一张表。但我们的线上需求通常是这样的：

我有一张非常大的表，行数超过十亿，需要进行拆分处理。假设拆分因子是512。 如果采用单node单数据库的分表方式，那其实这512个子表还是存在一个物理节点上，意义不大。 如果采用他们的sharding功能，就需要512个物理节点，也不现实。 面对这种需求，现有的proxy就不能很好地满足要求了。通常我们希望将512张子表均分在几个MySQL节点上，从而达到系统的横向扩展。

然而kingshard较好地实现了这种典型的需求。简单来说，kingshard的分表方案采用两级映射的方式：

1.kingshard将该表分成512张子表，例如：test_0000,test_0001,...
test_511。
2.将shardKey通过hash或range方式定位到其要操作的记录在哪张子表上。
3.子表落在哪个node上通过配置文件设置。

1.2  sharding支持的操作

目前kingshard sharding支持insert, delete, select, update和replace语句, 所有这五类操作都支持跨子表。但写操作仅支持单node上的跨子表，select操作则可以跨node，跨子表。

1.3  sharding方式

1.3.1  range方式

基于整数范围划分来得到子表下标。该方式的优点：基于范围的查询或更新速度快，因为查询（或更新）的范围有可能落在同一张子表中。这样可以避免全部子表的查询（更新）。缺点：数据热点问题。因为在一段时间内整个集群的写压力都会落在一张子表上。此时整个mysql集群的写能力受限于单台mysql server的性能。并且，当正在集中写的mysql 节点如果宕机的话，整个mysql集群处于不可写状态。基于range方式的分表字段类型受限。

1.3.2  hash方式

kingshard采用（shardKey%子表个数）的方式得到子表下标。优点：数据分布均匀，写压力会比较平均地落在后端的每个MySQL节点上，整个集群的写性能不会受限于单个MySQL节点。并且当某个分片节点宕机，只会影响到写入该节点的请求，其他节点的写入请求不受影响。分表字段类型不受限。因为任何一个类型的分表字段，都可以通过一个hash函数计算得到一个整数。缺点：基于范围的查询或更新，都需要将请求发送到全部子表，对性能有一定影响。但如果不是基于范围的查询或更新，则性能不会受到影响。

1.3.3  按时间分表

kingshard中的分表字段支持MySQL中三种类型的时间格式

• date类型，格式：YYYY-MM-DD，例如:2016-03-04,注意：2016-3-04，2016-03-4，2016-3-4等格式kingshard都是不支持的。
• datetime类型，格式：YYYY-MM-DD HH:MM:SS，例如:2016-03-04 13:23:43,注意：2016-3-04 13:23:43，2016-03-4 13:23:43，2016-3-4 13:23:43等格式kingshard都是不支持的，必须严格按照规定的格式，kingshard才支持。
• timestamp类型，整数类型，例如：1457165568，对应的是：2016-3-5 16:12:48

1.4  sharding相关的配置介绍

在配置文件中，有关sharding设置是通过schema设置：

schema :
-
   nodes: [node1,node2]
   rules:
       default: node1
       shard:
       -   
           #分表所在的DB
   		         db : kingshard
   		         #分表名字
           table: test_shard_hash
           #sharding key
           key: id
           #子表分布的节点名字
           nodes: [node1, node2]
           #sharding类型
           type: hash
           #子表个数分布，表示[test_shard_hash_0000, test_shard_hash_0001, test_shard_hash_0002, test_shard_hash_003]在node1上。
           #[test_shard_hash_0004, test_shard_hash_0005, test_shard_hash_0006, test_shard_hash_007]在node2上
           locations: [4,4]

       -   
           #分表所在的DB
   		         db : kingshard
   		         #分表名字
           table: test_shard_range
           #sharding key
           key: id
           #sharding类型
           type: range
           #子表分布的节点名字
           nodes: [node1, node2]
           #子表个数分布，表示[test_shard_range_0000, test_shard_range_0001, test_shard_range_0002, test_shard_range_003]在node1上。
           #[test_shard_range_0004, test_shard_range_0005, test_shard_range_0006, test_shard_range_007]在node2上
           locations: [4,4]
           #每张子表的记录数。[0,10000)在test_shard_range_0000上，[10000,20000)在test_shard_range_0001上。....
           table_row_limit: 10000

一个kingshard实例只能有一个schemas，从上面的配置可以看出，schema可以分为三个部分：

1.db，表示这个schemas使用的数据库。

2.nodes，表示子表分布的节点名字。

3.rules，sharding规则。其中rules又可以分为两个部分：
	- default，默认分表规则。所有操作不在shard（default规则下面的规则）中的表的SQL语句都会发向该node。
	- hash，hash分表方式。
	- range，range分表方式

1.5  分表架构图

二、sharding方式演示

2.1  range分表方式演示

2.1.1  基础环境准备

分别在三个节点创建测试表：

建表语句：
CREATE TABLE `test_shard_range_0000` (
  `id` bigint(64) unsigned NOT NULL,
  `str` varchar(256) DEFAULT NULL,
  `f` double DEFAULT NULL,
  `e` enum('test1','test2') DEFAULT NULL,
  `u` tinyint(3) unsigned DEFAULT NULL,
  `i` tinyint(4) DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8

node1：
mysql> show tables;
+-----------------------+
| Tables_in_test        |
+-----------------------+
| test_shard_range_0000 |
| test_shard_range_0001 |
+-----------------------+
2 rows in set (0.00 sec)

node2：
mysql> show tables;
+-----------------------+
| Tables_in_test        |
+-----------------------+
| test_shard_range_0002 |
| test_shard_range_0003 |
+-----------------------+
2 rows in set (0.00 sec)

node3：
mysql> show tables;
+-----------------------+
| Tables_in_test        |
+-----------------------+
| test_shard_range_0004 |
| test_shard_range_0005 |
+-----------------------+
2 rows in set (0.00 sec)

kingshard配置及启动：

配置文件

addr : 0.0.0.0:7676
user :  tom
password : 123456
web_addr : 0.0.0.0:9797
web_user : admin
web_password : admin
log_path : /home/mysql/3309_kingshard/log/ks.log
log_level : debug
log_sql: on
allow_ips : 127.0.0.1
nodes :
-
    name : node1
    max_conns_limit : 32
    user :  tom
    password : 123456
    master : xxxxxx:3309
    down_after_noalive : 32
-
    name : node2
    max_conns_limit : 32
    user :  tom
    password : 123456
    master : xxxxxxx:3309
    down_after_noalive: 32
-
    name : node3
    max_conns_limit : 32
    user :  tom
    password : 123456
    master : xxxxxxx:3309
    down_after_noalive : 32
schema :
    nodes: [node1,node2,node3]
    default: node1
    shard:
#按hash分表
#    -
#        db : test
#        table: test_shard_hasg
#        key: id
#        nodes: [node1, node2, node3]
#        type: hash
#        locations: [2,2,2]
#
#按range分表
#    -
#        db : test
#        table: test_shard_range
#        key: id
#        type: range
#        nodes: [node1, node2, node3]
#        locations: [2,2,2]
#        table_row_limit: 10
#按year分表
    -
        db : test
        table: test_shard_year
        key: ctime
        nodes: [node1, node2, node3]
        type: date_year
        date_range: [2016,2017,2018]
#    -
#        db : kingshard
#        table: test_shard_month
#        key: dtime
#        type: date_month
#        nodes: [node1,node2]
#        date_range: [201603-201605,201609-201612]
#    -
#        db : kingshard
#        table: test_shard_day
#        key: mtime
#        type: date_day
#        nodes: [node1,node2]
#        date_range: [20160306-20160307,20160308-20160309]

注意：nodes的信息要和表信息还有schema里面的nodes信息对应，不然分表写入会有问题

启动

./bin/kingshard -config=./conf/ks.yaml

2.1.2  测试演示

在kingshard下面操作：
mysql> insert into test_shard_range(id,str,f,e,u,i) values(1,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(11,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(21,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(31,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(41,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(51,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

跨节点查询：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |    3 |
| 11 | flike | 3.14 | test2 |    2 |    3 |
| 21 | flike | 3.14 | test2 |    2 |    3 |
| 31 | flike | 3.14 | test2 |    2 |    3 |
| 41 | flike | 3.14 | test2 |    2 |    3 |
| 51 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
10 rows in set (0.00 sec)

跨节点更新：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |    3 |
| 11 | flike | 3.14 | test2 |    2 |    3 |
| 21 | flike | 3.14 | test2 |    2 |    3 |
| 22 | flike | 3.14 | test2 |    2 |    3 |
| 23 | flike | 3.14 | test2 |    2 |    3 |
| 31 | flike | 3.14 | test2 |    2 |    3 |
| 32 | flike | 3.14 | test2 |    2 |    3 |
| 33 | flike | 3.14 | test2 |    2 |    3 |
| 41 | flike | 3.14 | test2 |    2 |    3 |
| 51 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
10 rows in set (0.00 sec)

mysql> update test_shard_range set i=100;
Query OK, 10 rows affected (0.00 sec)

mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |  100 |
| 11 | flike | 3.14 | test2 |    2 |  100 |
| 21 | flike | 3.14 | test2 |    2 |  100 |
| 22 | flike | 3.14 | test2 |    2 |  100 |
| 23 | flike | 3.14 | test2 |    2 |  100 |
| 31 | flike | 3.14 | test2 |    2 |  100 |
| 32 | flike | 3.14 | test2 |    2 |  100 |
| 33 | flike | 3.14 | test2 |    2 |  100 |
| 41 | flike | 3.14 | test2 |    2 |  100 |
| 51 | flike | 3.14 | test2 |    2 |  100 |
+----+-------+------+-------+------+------+
10 rows in set (0.00 sec)

跨节点删除：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |  100 |
| 11 | flike | 3.14 | test2 |    2 |  100 |
| 21 | flike | 3.14 | test2 |    2 |  100 |
| 22 | flike | 3.14 | test2 |    2 |  100 |
| 23 | flike | 3.14 | test2 |    2 |  100 |
| 31 | flike | 3.14 | test2 |    2 |  100 |
| 32 | flike | 3.14 | test2 |    2 |  100 |
| 33 | flike | 3.14 | test2 |    2 |  100 |
| 41 | flike | 3.14 | test2 |    2 |  100 |
| 51 | flike | 3.14 | test2 |    2 |  100 |
+----+-------+------+-------+------+------+
10 rows in set (0.00 sec)

mysql> delete from test_shard_range where id >11 and id <51;
Query OK, 7 rows affected (0.00 sec)

mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |  100 |
| 11 | flike | 3.14 | test2 |    2 |  100 |
| 51 | flike | 3.14 | test2 |    2 |  100 |
+----+-------+------+-------+------+------+
3 rows in set (0.01 sec)

对子查询的支持：（不支持）
mysql> select * from test_shard_range where id in (select u from test_shard_range where id=1);
ERROR 1146 (42S02): Table 'test.test_shard_range' doesn't exist

在node1下查看：
mysql> select * from test_shard_range_0000;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

mysql> select * from test_shard_range_0001;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
| 11 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

在node2下查看：
mysql> select * from test_shard_range_0002;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
| 21 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
3 rows in set (0.00 sec)

mysql> select * from test_shard_range_0003;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
| 31 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
3 rows in set (0.00 sec)

在node3下查看：
mysql> select * from test_shard_range_0004;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
| 41 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

mysql> select * from test_shard_range_0005;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
| 51 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

结论：range分表更能比较好用，不过不支持子查询，如果支持子查询就好了

2.2  hash分表方式演示

插入操作：
mysql> insert into test_shard_range(id,str,f,e,u,i) values(1,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(2,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(3,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(4,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(5,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(6,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(7,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(8,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_range(id,str,f,e,u,i) values(9,"flike",3.14,'test2',2,3);
Query OK, 1 row affected (0.00 sec)

查询操作：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  6 | flike | 3.14 | test2 |    2 |    3 |
|  1 | flike | 3.14 | test2 |    2 |    3 |
|  7 | flike | 3.14 | test2 |    2 |    3 |
|  2 | flike | 3.14 | test2 |    2 |    3 |
|  8 | flike | 3.14 | test2 |    2 |    3 |
|  3 | flike | 3.14 | test2 |    2 |    3 |
|  9 | flike | 3.14 | test2 |    2 |    3 |
|  4 | flike | 3.14 | test2 |    2 |    3 |
|  5 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
9 rows in set (0.01 sec)

更新操作：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  6 | flike | 3.14 | test2 |    2 |    3 |
|  1 | flike | 3.14 | test2 |    2 |    3 |
|  7 | flike | 3.14 | test2 |    2 |    3 |
|  2 | flike | 3.14 | test2 |    2 |    3 |
|  8 | flike | 3.14 | test2 |    2 |    3 |
|  3 | flike | 3.14 | test2 |    2 |    3 |
|  9 | flike | 3.14 | test2 |    2 |    3 |
|  4 | flike | 3.14 | test2 |    2 |    3 |
|  5 | flike | 3.14 | test2 |    2 |    3 |
+----+-------+------+-------+------+------+
9 rows in set (0.01 sec)

mysql> update test_shard_range set u=10;
Query OK, 9 rows affected (0.00 sec)

mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  6 | flike | 3.14 | test2 |   10 |    3 |
|  1 | flike | 3.14 | test2 |   10 |    3 |
|  7 | flike | 3.14 | test2 |   10 |    3 |
|  2 | flike | 3.14 | test2 |   10 |    3 |
|  8 | flike | 3.14 | test2 |   10 |    3 |
|  3 | flike | 3.14 | test2 |   10 |    3 |
|  9 | flike | 3.14 | test2 |   10 |    3 |
|  4 | flike | 3.14 | test2 |   10 |    3 |
|  5 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
9 rows in set (0.00 sec)

distinct函数的支持：
mysql> select * from test_shard_range;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  6 | flike | 3.14 | test2 |   10 |    3 |
|  1 | flike | 3.14 | test2 |   10 |    3 |
|  7 | flike | 3.14 | test2 |   10 |    3 |
|  2 | flike | 3.14 | test2 |   10 |    3 |
|  8 | flike | 3.14 | test2 |   10 |    3 |
|  3 | flike | 3.14 | test2 |   10 |    3 |
|  9 | flike | 3.14 | test2 |   10 |    3 |
|  4 | flike | 3.14 | test2 |   10 |    3 |
|  5 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
9 rows in set (0.00 sec)

mysql> select count(distinct id) as tom from test_shard_range;
+-----+
| tom |
+-----+
|   9 |
+-----+
1 row in set (0.00 sec)

mysql> select count(distinct u) as tom from test_shard_range;
+-----+
| tom |
+-----+
|   6 |
+-----+
1 row in set (0.00 sec)

mysql> select count(distinct i) as tom from test_shard_range;
+-----+
| tom |
+-----+
|   6 |
+-----+
1 row in set (0.00 sec)

group by的支持：
mysql> select * from test_shard_range where id > 0 group by u limit 1;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  5 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

mysql> select * from test_shard_range where id > 0 group by u limit 1;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  3 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

mysql> select * from test_shard_range where id > 0 group by u limit 1;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.00 sec)

mysql> select * from test_shard_range where id > 0 group by u limit 1;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  5 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
1 row in set (0.01 sec)

order by的支持：
mysql> select * from test_shard_range where id > 0 order by id;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |   10 |    3 |
|  2 | flike | 3.14 | test2 |   10 |    3 |
|  3 | flike | 3.14 | test2 |   10 |    3 |
|  4 | flike | 3.14 | test2 |   10 |    3 |
|  5 | flike | 3.14 | test2 |   10 |    3 |
|  6 | flike | 3.14 | test2 |   10 |    3 |
|  7 | flike | 3.14 | test2 |   10 |    3 |
|  8 | flike | 3.14 | test2 |   10 |    3 |
|  9 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
9 rows in set (0.00 sec)

mysql> select * from test_shard_range where id > 0 order by id limit 5;
+----+-------+------+-------+------+------+
| id | str   | f    | e     | u    | i    |
+----+-------+------+-------+------+------+
|  1 | flike | 3.14 | test2 |   10 |    3 |
|  2 | flike | 3.14 | test2 |   10 |    3 |
|  3 | flike | 3.14 | test2 |   10 |    3 |
|  4 | flike | 3.14 | test2 |   10 |    3 |
|  5 | flike | 3.14 | test2 |   10 |    3 |
+----+-------+------+-------+------+------+
5 rows in set (0.00 sec)

注意：

1、当更新的记录落在不同的子表，kingshard会以非事务的方式将更新操作发送到多个node上。

2、不支持distinct函数。

3、跨节点的group by可能每次出来的结果都不一样。

2.3  date分表演示

创建测试表：
CREATE TABLE `test_shard_year_2018` (
  `id` int(10) NOT NULL,
  `name` varchar(40) DEFAULT NULL,
  `ctime` datetime DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8

插入数据：
mysql> insert into test_shard_year(id,name,ctime) values(12,"hello","2016-02-22 13:23:45");
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_year(id,name,ctime) values(12,"hello","2017-02-22 13:23:45");
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_year(id,name,ctime) values(12,"hello","2018-02-22 13:23:45");
Query OK, 1 row affected (0.00 sec)

mysql> insert into test_shard_year(id,name,ctime) values(12,"hello","2019-02-22 13:23:45");
ERROR 1146 (42S02): Table 'test.test_shard_year_2019' doesn't exist

查看数据：
mysql> select * from test_shard_year;
+----+-------+---------------------+
| id | name  | ctime               |
+----+-------+---------------------+
| 12 | hello | 2016-02-22 13:23:45 |
| 12 | hello | 2017-02-22 13:23:45 |
| 12 | hello | 2018-02-22 13:23:45 |
+----+-------+---------------------+
3 rows in set (0.01 sec)

mysql> /*node1*/select * from test_shard_year_2016;
+----+-------+---------------------+
| id | name  | ctime               |
+----+-------+---------------------+
| 12 | hello | 2016-02-22 13:23:45 |
+----+-------+---------------------+
1 row in set (0.00 sec)

mysql> /*node1*/select * from test_shard_year_2017;
ERROR 1146 (42S02): Table 'test.test_shard_year_2017' doesn't exist
mysql> /*node2*/select * from test_shard_year_2017;
+----+-------+---------------------+
| id | name  | ctime               |
+----+-------+---------------------+
| 12 | hello | 2017-02-22 13:23:45 |
+----+-------+---------------------+
1 row in set (0.00 sec)

mysql> /*node3*/select * from test_shard_year_2018;
+----+-------+---------------------+
| id | name  | ctime               |
+----+-------+---------------------+
| 12 | hello | 2018-02-22 13:23:45 |
+----+-------+---------------------+
1 row in set (0.00 sec)

mysql> /*node3*/select * from test_shard_year;
ERROR 1146 (42S02): Table 'test.test_shard_year' doesn't exist

结论：date分表功能可用。

参考地址：https://github.com/flike/kingshard/

三、kingshard性能测试、优化

https://github.com/flike/kingshard/blob/master/doc/KingDoc/kingshard_performance_test.md

https://github.com/flike/kingshard/blob/master/doc/KingDoc/kingshard_performance_profiling.md

 

来源：oschina

链接：https://my.oschina.net/u/3023401/blog/1808333