MariaDB ColumnStore专为大数据扩展而设计,可处理PB级数据,线性可伸缩性和出色的性能,并能对分析查询进行实时响应。 它利用列式存储,压缩,即时投影以及水平和垂直分区的I / O优势在分析大型数据集时提供了出色的性能。(这是官方的介绍,https://mariadb.com/kb/en/mariadb-columnstore/)
本人自研发了一个mysql -> mariadb columnStore同步工具,同时支持全量同步和增量同步模式,github地址:https://github.com/yutianyong125/mcs_etl
接下来演示一下同步过程
下载该项目
git clone https://github.com/yutianyong125/mcs_etl.git
cd mcs_etl
启动mysql,方便起见,这里使用docker,仅测试,不考虑数据挂载问题,这里我把secure-file-priv配置的目录挂载出来,是为了导入数据的时候用到,另需要指定使用自定义的mysql配置文件
vim /tmp/conf.d/my.cnf
# 保存为以下配置
[mysqld]
log_bin = mysql_bin # 开启binlog日志并指定binlog日志命名前缀
binlog_format=ROW # 指定binlog格式
server_id = 1 # 指定master server_id
secure_file_priv=/tmp/mysql-files # SELECT INTO OUTFILE 语句需要开启这个配置
mkdir /tmp/mysql-files # 先创建好存放导出数据的文件夹
docker run --rm -d --name mysql -v /tmp/conf.d:/etc/mysql/conf.d -v /tmp/mysql-files:/tmp/mysql-files -e MYSQL_ROOT_PASSWORD=123456 -p 3306:3306 mysql:5.7 --character-set-server=utf8mb4 --collation-server=utf8mb4_unicode_ci --secure-file-priv=/tmp/mysql-files
使用存储过程添加测试数据
-- 创建test库
create database test;
use test;
-- 创建user表
CREATE TABLE user(
id INT NOT NULL AUTO_INCREMENT,
uname VARCHAR(20) NOT NULL,
sex VARCHAR(5) NOT NULL,
score INT NOT NULL,
copy_id INT NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=INNODB CHARSET=utf8;
-- 存储过程插入10万条数据
DROP PROCEDURE IF EXISTS add_user;
DELIMITER //
create PROCEDURE add_user(in num INT)
BEGIN
DECLARE rowid INT DEFAULT 0;
DECLARE firstname CHAR(1);
DECLARE name1 CHAR(1);
DECLARE name2 CHAR(1);
DECLARE sex CHAR(1);
DECLARE score CHAR(2);
DECLARE uname CHAR(3);
WHILE rowid < num DO
SET firstname = SUBSTRING('赵钱孙李周吴郑王林杨柳刘孙陈江阮侯邹高彭徐',FLOOR(1+21*RAND()),1);
SET name1 = SUBSTRING('一二三四五六七八九十甲乙丙丁静景京晶名明铭敏闵民军君俊骏天田甜兲恬益依成城诚立莉力黎励',ROUND(1+43*RAND()),1);
SET name2 = SUBSTRING('一二三四五六七八九十甲乙丙丁静景京晶名明铭敏闵民军君俊骏天田甜兲恬益依成城诚立莉力黎励',ROUND(1+43*RAND()),1);
SET sex=FLOOR(0 + (RAND() * 2));
SET score= FLOOR(40 + (RAND() *60));
SET rowid = rowid + 1;
SET uname = CONCAT(firstname,name1,name2);
insert INTO user (uname,sex,score,copy_id) VALUES (uname,sex,score,rowid);
END WHILE;
END //
DELIMITER ;
call add_user(100000);
-- 创建test1库
create database test1;
use test1;
-- 创建表t1
CREATE TABLE `t` (
`id` int(11) NOT NULL,
`a` int(11) DEFAULT NULL,
`b` int(11) DEFAULT NULL,
PRIMARY KEY (`id`),
KEY `a` (`a`),
KEY `b` (`b`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
-- 存储过程插入数据
delimiter //
create procedure idata()
begin
declare i int;
set i=1;
while(i<=100000)do
insert into t values(i, i, i);
set i=i+1;
end while;
end //
delimiter ;
call idata();
启动mariadb columnStore
docker run --rm -d --name mcs -eMARIADB_ROOT_PASSWORD=123456 -p 3307:3306 mariadb/columnstore:1.2.3
修改项目配置文件, conf/etl.toml
# 增量ETL配置段
[IncrementEtl]
StartFile = "" # 开始同步的binlog文件
StartPosition = 0 # 开始同步的binlog位置
ServerId = 100 # 同步slave的serverId,只要不与master serverId相同即可
# 全量ETL配置段
[FullEtl]
# mysql数据导出存放文件夹
OutFileDir = "/tmp/mysql-files/"
# 全量ETL规则,Schema: 数据库, Tables: table数组
# eg: 导出整个库的表 tables=["*"],导出特定的表 tables=["test1", "test2"]
[[Rule]]
Schema = "test"
Tables = ["user"]
[[Rule]]
Schema = "test1"
Tables = ["*"]
# mysql 数据库配置
[Source]
Host = "127.0.0.1"
Port = 3306
User = "root"
Pwd = "123456"
# mariadb columnStore 数据库配置
[Target]
Host = "127.0.0.1"
Port = 3307
User = "root"
Pwd = "123456"
执行二进制文件,指定全量同步模式
./mcs_etl -model full
成功输出如下,因为是多协程异步执行任务的,所以总耗时可能与etl时长最大的表相差无几
同步`test1`.`t` 耗时 2.025725559s
同步`test`.`user` 耗时 4.358782152s
fullEtl 耗时 4.373473788s
增量方式进行同步步骤如下:
mysql运行 show master status 查看当前binlog日志文件和Position,修改 conf/etl.toml 配置文件中的IncrementEtl配置段
# 增量ETL配置段
[IncrementEtl]
StartFile = "mysql_bin.000004" # 开始同步的binlog文件
StartPosition = 154 # 开始同步的binlog位置
ServerId = 100 # 同步slave的serverId,只要不与master serverId相同即可
执行二进制文件,指定增量同步模式,此时会进入循环监听等待消费状态
./mcs_etl -model increment
接下来在mysql做一些增删改操作
update `test`.`user` set sex = 0 where id = 1;
insert into `test1`.`t` values (100001, 100001, 100001);
delete from `test1`.`t` where id = 100001;
控制台输出日志如下
binlog恢复的语句:
update `test`.`user` set `id` = '1',`uname` = '邹二君',`sex` = '0',`score` = '59',`copy_id` = '1' where `id` = '1' and `uname` = '邹二君' and `sex` = '1' and `score` = '59' and `copy_id` = '1'
==>
兼容处理转换后的语句:
update `test`.`user` set `id` = '1',`uname` = '邹二君',`sex` = '0',`score` = '59',`copy_id` = '1' where `id` = '1' and `uname` = '邹二君' and `sex` = '1' and `score` = '59' and `copy_id` = '1'
执行结果:
执行成功
binlog恢复的语句:
insert into `test1`.`t` (`id`,`a`,`b`) values ('100001','100001','100001')
==>
兼容处理转换后的语句:
insert into `test1`.`t` (`id`,`a`,`b`) values ('100001','100001','100001')
执行结果:
执行成功
binlog恢复的语句:
delete from `test1`.`t` where `id` = '100001' and `a` = '100001' and `b` = '100001'
==>
兼容处理转换后的语句:
delete from `test1`.`t` where `id` = '100001' and `a` = '100001' and `b` = '100001'
执行结果:
执行成功