About databases and tables¶
Databases and tables are at the core of any Database Management System. ClickHouse® is no different. In this article we will look at how to create and work with databases and tables in Aiven for ClickHouse.
Our managed service Aiven for ClickHouse has some minor differences comparing to “classical” ClickHouse. These difference are needed to ensure security, stability and proper replication of data.
Creating databases in an Aiven for ClickHouse service can only be done via the Aiven platform; the admin user is not allowed to create databases directly for security and reliability reasons. However, you can create a new database through the web interface of Aiven console:
Log in to Aiven console and select your service.
On the service page open the tab Databases & Tables.
Enter your database name in Create a new database and click on Create database.
On our side we enable necessary customizations and run secondary queries to grant access to the admin user.
If you try adding a new database to Aiven for ClickHouse through the command line, you’ll encounter an exception
Not enough privileges.(ACCESS_DENIED). Please use the Aiven web interface to add a new database.
Tables can be added with an SQL query, either with the help of the web query editor or with the cli. In both cases the SQL query will look the same. For example, below is a query to add a new table
transactions database. To keep it simple this example has unrealistically small amount of columns:
CREATE TABLE transactions.expenses ( Title String, Date DateTime, UserID UInt64, Amount UInt32 ) ENGINE = ReplicatedMergeTree ORDER BY Date;
Part of the table definition includes a targeted table engine. The full list of supported table engines in Aiven for ClickHouse can be found in this article. Aiven for ClickHouse uses “Replicated” variants of table engines to ensure high availability. Even if you select
MergeTree engine, we will automatically use the replicated variant on our side.
Additionally, Aiven for ClickHouse takes care of running queries in a distributed mode affecting entire cluster. In the “classical” ClickHouse the queries
DROP will only affect the server where they are run. In contrast, we ensure the proper distribution across all cluster machines behind the scenes. You don’t need to remember using
ON CLUSTER for every query.
With this knowledge try out an example dataset described over here.