# Treasure Data

The *Treasure Data* (TD) output plugin lets you flush your records into the [Treasure Data](https://treasuredata.com) cloud service.

## Configuration parameters

The plugin supports the following configuration parameters:

| Key        | Description                                                                                                                                                     | Default |
| ---------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
| `api`      | The Treasure Data API key. To obtain it, log into the [Console](https://console.treasuredata.com/users/sign_in) and in the API keys box, copy the API key hash. | *none*  |
| `database` | Specify the name of your target database.                                                                                                                       | *none*  |
| `region`   | Set the service region. Allowed values: `US`, `JP`.                                                                                                             | `US`    |
| `Region`   | Classic-mode spelling of `region`. Allowed values: `us`, `jp`.                                                                                                  | `US`    |
| `table`    | Specify the name of your target table where the records will be stored.                                                                                         | *none*  |
| `workers`  | The number of [workers](https://docs.fluentbit.io/manual/administration/multithreading#outputs) to perform flush operations for this output.                    | `0`     |

## Get started

To start inserting records into Treasure Data, run the plugin from the command line or through the configuration file.

### Command line

You can run the plugin from the command line, but it exposes your API key. Using a configuration file is recommended.

```shell
fluent-bit -i cpu -o td -p API="abc" -p Database="fluentbit" -p Table="cpu_samples"
```

### Configuration file

In your main configuration file append the following:

{% tabs %}
{% tab title="fluent-bit.yaml" %}

```yaml
pipeline:
  inputs:
    - name: cpu
      tag: my_cpu

  outputs:
    - name: td
      match: '*'
      api: 'XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX'
      database: fluentbit
      table: cpu_samples
```

{% endtab %}

{% tab title="fluent-bit.conf" %}

```
[INPUT]
  Name cpu
  Tag  my_cpu

[OUTPUT]
  Name     td
  Match    *
  API      XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
  Database fluentbit
  Table    cpu_samples
```

{% endtab %}
{% endtabs %}