> ## Documentation Index
> Fetch the complete documentation index at: https://docs.btsscorp.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect your clouds

> Point Koala at your OCI, AWS, Azure, and GCP cost exports

Each cloud connection tells Koala **where your FOCUS cost export lives and how to read it**. You can add more than one configuration per cloud, mark one as the default, and turn on automatic scheduled ingestion. Credentials are **encrypted at rest** and shown only as masked hints.

Set connections up in **Settings → Cloud accounts**. Each provider is below.

## Oracle Cloud (OCI)

Koala reads FOCUS reports from an **OCI Object Storage** bucket using an API-key signing key.

| Field                        | What it is                                                          |
| ---------------------------- | ------------------------------------------------------------------- |
| **Tenancy OCID / User OCID** | The identities the API key belongs to.                              |
| **Fingerprint**              | Your API signing key's fingerprint.                                 |
| **Private key (PEM)**        | The signing key content.                                            |
| **Region**                   | e.g. `us-ashburn-1`.                                                |
| **Bucket & FOCUS prefix**    | The bucket and path (default `FOCUS Reports/`) holding the exports. |

<Note>OCI connections authenticate with an **API key** (instance principals aren't used).</Note>

## Amazon Web Services (AWS)

Koala reads FOCUS exports from an **S3 bucket**.

| Field                         | What it is                                                                                                |
| ----------------------------- | --------------------------------------------------------------------------------------------------------- |
| **S3 bucket, prefix, region** | Where the exports live and which region.                                                                  |
| **File pattern**              | Which objects to match (default `*.csv.gz`).                                                              |
| **Credentials**               | An access key + secret (optionally a session token), **or** an **IAM role ARN** for cross-account access. |

## Microsoft Azure

Koala reads FOCUS exports from an **Azure Blob** container.

| Field                           | What it is                                                  |
| ------------------------------- | ----------------------------------------------------------- |
| **Storage account & container** | Where the exports live.                                     |
| **Authentication**              | A **service principal** *or* a **SAS token** — provide one. |

## Google Cloud (GCP)

Koala reads billing data from a **BigQuery billing export**.

| Field                          | What it is                                                      |
| ------------------------------ | --------------------------------------------------------------- |
| **Project ID & dataset**       | The project and BigQuery dataset holding the export.            |
| **Table pattern**              | The export tables to match (default `gcp_billing_export_v1_*`). |
| **Location**                   | The dataset location (default `US`).                            |
| **Service account key (JSON)** | Optional — if omitted, application default credentials apply.   |

## Defaults and scheduling

Every provider supports the same two options:

* **Default configuration** — the one Koala uses when you don't pick another.
* **Automatic ingestion** — turn on a schedule (a cron expression) and a **look-back window** (how many days of files to re-check on each run), and Koala keeps your data current on its own.

## Next steps

<Card title="Ingestion" icon="file-import" href="/finops/connections/ingestion">
  Upload files, run a load on demand, or schedule automatic ingestion.
</Card>