> For the complete documentation index, see [llms.txt](https://docs.marpledata.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.marpledata.com/docs/marple-insight/setup-and-configuration/data-connection/connection-configuration.md).

# Connection configuration

Wire up your connection by following the following steps (detailed explanation below):

1. Choose database type
2. Connect to your database
3. Define datasets
4. Write the required queries

To establish a connection, navigate to **Settings → Connection**.

<figure><img src="/files/JSSJ0FjgqxeifRTYbUUf" alt="" width="563"><figcaption></figcaption></figure>

## Step 1: Choose database type

In the **Overview** tab, you can change the type of database you want to connect to.

You can also export the current connection setup (including all the queries), or import connection settings from a JSON.

To get some guidance, you can start by uploading these test queries (written for Postgres databases):

{% file src="/files/uMRum57Ljb0XhSVHXc1G" %}

<figure><img src="/files/TIdi8Ns3IySn6TWbj9Ns" alt=""><figcaption><p>Connection overview</p></figcaption></figure>

Read more:

{% content-ref url="/pages/27I0fxw8rW4lZlpYTyXW" %}
[Supported Databases](/docs/marple-insight/setup-and-configuration/data-connection/supported-databases.md)
{% endcontent-ref %}

## Step 2: Connect to your database

Before configuring a connection, ensure that you have configured all necessary credentials. The required values differ between database types.&#x20;

After selecting the database type in the **Overview** tab, move to **Connection** to fill in the credentials.

Use "Test Connection" to make sure everything works. It should give a green checkmark (✅)

<figure><img src="/files/hFgU6wYvXlGad3BqwuDO" alt=""><figcaption><p>Connection credentials</p></figcaption></figure>

## Step 3: Define datasets

Marple Insight has a "dataset-oriented" approach.\
Start by defining 1 or more dataset metadata keys that define a dataset uniquely. Each dataset will shown in the library as a separate entry that can be analysed.

A dataset can be for example one flight test, race data for 1 car, ...

In the **Metadata** tab, you can define all dataset metadata that is shown in Marpe.

Mark the dataset identifiers as "filter". Those values can later be used in the queries, e.g. as *dataset.dataset\_id*.

<figure><img src="/files/lxakSt011pbr143kyydb" alt="Metadata as dataset identifiers"><figcaption></figcaption></figure>

## Step 4: Queries

Next, within the Marple Insight application, you'll need to define a set of queries that allow Marple to interpret and retrieve your data.&#x20;

There are **required queries**—such as those for datasets, signal lists, and time-series data—as well as **optional queries**, like segment-based queries.

Make sure to wire up all required queries to enable Marple Insight on your data.

{% content-ref url="/pages/euN2F0K10q9TWLyBBgJy" %}
[Required Queries](/docs/marple-insight/setup-and-configuration/data-connection/connection-configuration/required-queries.md)
{% endcontent-ref %}

{% content-ref url="/pages/eoWSYKXGgjR9aahD9zhz" %}
[Optional Queries](/docs/marple-insight/setup-and-configuration/data-connection/connection-configuration/optional-queries.md)
{% endcontent-ref %}

These queries are written using the Jinja templating language, which makes them dynamic and adaptable to different database structures.

{% content-ref url="/pages/B4n5EcAPkbCV7AXG7fhA" %}
[Templating](/docs/marple-insight/setup-and-configuration/data-connection/connection-configuration/templating.md)
{% endcontent-ref %}

## Need help?

Our team has a lot of expertise in setting up Marple Insight for customers. Get in touch if you have any questions:

{% content-ref url="/pages/8SkXB1SRqj5MoqRQzPMa" %}
[Feedback](/docs/marple-insight/welcome/feedback.md)
{% endcontent-ref %}