Clickhouse deployment and basic usage 1

1. Deployment structure

stay Clickhouse in , In general, there are 2 There are two ways to deploy ,3 Method of use ：

clickhouse Deployment structure .png

Deployment way ：

• Deployment way 1: Deploy a single node Clickhouse colony , For example, the cluster used by Xiaobai .
• Deployment way 2: Deploy a multi node cluster , such as Tom and Jim Cluster used

Usage mode ：

• Usage mode 1： Single node cluster , Single node execution SQL Direct manipulation . For example, Xiaobai in the above figure (sql No middle on cluster sentence ）
• Usage mode 2:： Multi node cluster , And how to use it 1 equally , Like the one above Tom(sql No middle on cluster sentence ） This way, , If the user is connected to Different nodes What you see The data may be different Of .
• Usage mode 3： Multi node cluster , user Cluster operation , Like the one above Sam and Jim Cluster operation , SQL Statement needs to be added on cluster XXXX, XXX Indicates the cluster name .

The following is a brief introduction to the basic functions of cluster operation and single node operation SQL.

2. Cluster operation Demo SQL

• 2.1 Create a database

create database db_test on cluster default_cluster

• 2.2. Delete a database

drop database db_test on cluster default_cluster

• 2.3. Create a table

CREATE TABLE db_test.tbl_replac_merge_test on cluster default_cluster
(
 id String, 
 code String, 
 create_time DateTime
)
ENGINE = ReplacingMergeTree()
PARTITION BY toYYYYMM(create_time)
PRIMARY KEY id
ORDER BY (id, code);

• 2.4. Delete a table

drop table db_test.tbl_replac_merge_test on cluster default_cluster

3. Single node operation Demo SQL

• 3.1. Create a database

create database db_test;

• 3.2. Delete a database

drop database db_test;

• 3.3. Create a table

CREATE TABLE db_test.tbl_replac_merge_test
(
`id` String, 
`code` String, 
`create_time` DateTime
)
ENGINE = ReplacingMergeTree()
PARTITION BY toYYYYMM(create_time)
PRIMARY KEY id
ORDER BY (id, code);

• 3.4. Delete a database table

drop table db_tetbl_replac_merge_test

4. How to create distributed tables and local tables

• 4.1 Create database

CREATE DATABASE db_100 on cluster default_cluster;

• 4.2 Create local table

CREATE TABLE db_100.tb_1_local on cluster default_cluster
(
    `c1` Int64,
    `c2` String,
    `_sign` Int8,
    `_version` UInt64
)
ENGINE = ReplicatedReplacingMergeTree('/clickhouse/tables/5a165fe5-ab38-4bf2-a6ee-5124a65ed1aa/{shard}', '{replica}', _version)
ORDER BY (c1, c2)
SETTINGS index_granularity = 8192

• 4.3 Create a distributed table

CREATE TABLE db_100.tb_1 on cluster default_cluster
(
    `c1` Int64,
    `c2` String,
    `_sign` Int8,
    `_version` UInt64
)
ENGINE = Distributed('default_cluster', 'db_1', 'tb_1_local', cityHash64(toString((c1, c2))))

---

more Clickhouse Table creation document ：

https://clickhouse.tech/docs/en/sql-reference/statements/create/table/