> ## Documentation Index
> Fetch the complete documentation index at: https://private-7c7dfe99-fix-nav-issues.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Creating tables in ClickHouse

> Learn about Creating Tables in ClickHouse

Like most databases, ClickHouse logically groups tables into **databases**. Use the `CREATE DATABASE` command to create a new database in ClickHouse:

```sql theme={null}
CREATE DATABASE IF NOT EXISTS helloworld
```

Similarly, use `CREATE TABLE` to define a new table. If you don't specify the database name, the table will be in the
`default` database.

The following table named `my_first_table` is created in the `helloworld` database:

```sql theme={null}
CREATE TABLE helloworld.my_first_table
(
    user_id UInt32,
    message String,
    timestamp DateTime,
    metric Float32
)
ENGINE = MergeTree()
PRIMARY KEY (user_id, timestamp)
```

In the example above, `my_first_table` is a `MergeTree` table with four columns:

* `user_id`:  a 32-bit unsigned integer
* `message`: a `String` data type, which replaces types like `VARCHAR`, `BLOB`, `CLOB` and others from other database systems
* `timestamp`: a `DateTime` value, which represents an instant in time
* `metric`: a 32-bit floating point number

<Note>
  The table engine determines:

  * How and where the data is stored
  * Which queries are supported
  * Whether or not the data is replicated

  There are many engines to choose from, but for a simple table on a single-node ClickHouse server, [MergeTree](/reference/engines/table-engines/mergetree-family/mergetree) is your likely choice.
</Note>

<h2 id="a-brief-intro-to-primary-keys">
  A brief intro to primary keys
</h2>

Before you go any further, it is important to understand how primary keys work in ClickHouse (the implementation
of primary keys might seem unexpected!):

* primary keys in ClickHouse are ***not unique*** for each row in a table

The primary key of a ClickHouse table determines how the data is sorted when written to disk. Every 8,192 rows or 10MB of
data (referred to as the **index granularity**) creates an entry in the primary key index file. This granularity concept
creates a **sparse index** that can easily fit in memory, and the granules represent a stripe of the smallest amount of
column data that gets processed during `SELECT` queries.

The primary key can be defined using the `PRIMARY KEY` parameter. If you define a table without a `PRIMARY KEY` specified,
then the key becomes the tuple specified in the `ORDER BY` clause. If you specify both a `PRIMARY KEY` and an `ORDER BY`, the primary key must be a prefix of the sort order.

The primary key is also the sorting key, which is a tuple of `(user_id, timestamp)`.  Therefore, the data stored in each
column file will be sorted by `user_id`, then `timestamp`.

<Tip>
  For more details, check out the [Modeling Data training module](https://learn.clickhouse.com/visitor_catalog_class/show/1328860/?utm_source=clickhouse\&utm_medium=docs) in ClickHouse Academy.
</Tip>