clickhouse常用SQL语句，查询、建表、数据复制迁移、删除等
坚持是一种态度
于 2022-02-18 17:58:35 发布
1646
收藏 6
分类专栏： 数据库及存储技术 大数据开发 文章标签： sql 数据库 database
版权
数据库及存储技术
同时被 2 个专栏收录
38 篇文章2 订阅
订阅专栏
大数据开发
12 篇文章0 订阅
订阅专栏
文章目录
1. 查询 数据库容量
2. 查看所有表
3. 常用查询
4. MySQL导入到ClickHouse
5. clickhouse表数据迁移到新的分区表
1. 查询 数据库容量
select
sum(rows) as "总行数",
formatReadableSize(sum(data_uncompressed_bytes)) as "原始大小",
formatReadableSize(sum(data_compressed_bytes)) as "压缩大小",
round(sum(data_compressed_bytes) / sum(data_uncompressed_bytes) * 100, 0) "压缩率"
from system.parts;
2. 查看所有表
SELECT database, table, partition, name, active
FROM system.parts
WHERE table = 'table_name';
3. 常用查询
-- 列出数据库列表
show databases;
-- 列出数据库中表列表
show tables;
-- 创建数据库
create database test;
-- 删除一个表
drop table if exists test.t1;
-- 创建一个表
DROP TABLE visit_record_partition_month;
CREATE TABLE default.visit_record_partition_month ( `CREATE_DATE` DateTime,
`PLATFORM_TYPE` String, `TERMINAL_UNIQUE_ID` String,`SYSTEM_UNIQUE_ID` String,
`USER_UNIQUE_ID` String,`IP` String,`REFERER` String,`URL` String,`TITLE` String,
`COOKIE` String,`SCOOKIE` String,`COUNTRY` String,`PROVINCE` String,
`CITY` String,`SEARCH_ENGINE` String,`SEARCH_KEY` String,`SOURCE_TYPE` String,
`SOURCE_HOST` String,`OS` String,`CLIENT` String,`IS_PC` String,`RESOLUTION` String,
`COLORDEPTH` String,`LANGUAGE` String,`TYPE_CODE` String,`COLUMN_CLASS_CODE` String,
`PUBLIC_INFO_ATTRIBUTE` String,`CLASS_CODE` String, `COLUMN_TYPE` String,
`COLUMN_NAME` String,`LABELS` String,`COLUMN_ID` Int64 DEFAULT 0,`UNIT_ID` Int64 DEFAULT 0,
`CAT_ID` Int64 DEFAULT 0) ENGINE = MergeTree
PARTITION BY toYYYYMM(CREATE_DATE) ORDER BY CREATE_DATE;
-- 插入测试数据
insert into default.visit_record_partition_month (CREATE_DATE,PLATFORM_TYPE, CITY) values ('2021-12-12 12:11:11',1, 'abc'), ('2022-01-12 12:11:11',2, 'bbbb');
-- 查询
select * from default.visit_record_partition_month;
-- 分区名可以用下语句查询
SELECT table, partition, path FROM system.parts WHERE table = 'visit_record_partition_month';
-- 删除分区
alter table visit_record drop partition '202104';
4. MySQL导入到ClickHouse
进入clickhouse client命令界面操作，其他数据库也类似，建立远程连接即可
clickhouse-client -m -h 127.0.0.1 --password Lonsun#0717
1
执行转库SQL，经测试，256秒转换4837万数据，平均18.9万条每秒，平均94.33M每秒
INSERT INTO visit_record(CREATE_DATE,PLATFORM_TYPE,TERMINAL_UNIQUE_ID,SYSTEM_UNIQUE_ID,USER_UNIQUE_ID,IP,REFERER,URL,
TITLE,COOKIE,SCOOKIE,COUNTRY,PROVINCE,CITY,SEARCH_ENGINE,SEARCH_KEY,SOURCE_TYPE,SOURCE_HOST,OS,CLIENT,IS_PC,RESOLUTION,COLORDEPTH,LANGUAGE)
select CREATE_DATE, 'EX',ifNull(IP,S_COOKIE),concat('sz_gova_', SITE_ID),ifNull(MEMBER_ID,''),ifNull(IP,''),ifNull(REFERER,''),ifNull(URL,''),ifNull(TITLE,''),ifNull(COOKIE,''),
ifNull(S_COOKIE,''),ifNull(COUNTRY,''),ifNull(PROVINCE,''),ifNull(CITY,''),ifNull(SEARCH_ENGINE,''),ifNull(SEARCH_KEY,''),ifNull(SOURCE_TYPE,''),ifNull(SOURCE_HOST,''),ifNull(OS,''),
ifNull(CLIENT,''),ifNull(IS_PC,''),ifNull(RESOLUTION,''),ifNull(COLOR_DEPTH,''),ifNull(LANGUAGE,'')
from mysql('xx.xxx.xx.xxx:3306', 'ex9_1', 'cms_site_chart_main', 'root', 'xylx1.t!@#') where ID > 484794271 order by ID;
1
2
3
4
5
6
5. clickhouse表数据迁移到新的分区表
INSERT INTO visit_record_partition_month SELECT * FROM visit_record;
-- 导入后，查看该表的现有分区和数据情况
SELECT * FROM system.parts WHERE table = 'visit_record_partition_month';
1
2
3
4
————————————————
版权声明：本文为CSDN博主「坚持是一种态度」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/u010882234/article/details/123008705