Data Version Control (DVC)

Outpost integrates with DVC to manage large files, datasets, and machine learning models. It seamlessly works with your existing Outpost Git repositories, providing remote object storage managed by DVC, file diffing for DVC-tracked files hosted on Outpost Storage or S3-compatible storage, and data pipeline visualization.

How Does Outpost Integrate with DVC?

Outpost Storage

Each repository on Outpost automatically gets a remote object storage setup with 100 GB of free space. This storage, managed by DVC, is easily configurable on any machine. Using DVC pointer files (.dvc) and the dvc.lock file, Outpost parses the storage and displays DVC-tracked files under the Files tab.

Learn more about Outpost Storage.

External Storage Buckets

Outpost supports DVC data stored on various external storage solutions, including AWS S3, Google Cloud Storage, and Azure Blob Storage.

Discover how to configure your external bucket.

Visualize DVC Pipelines

Outpost interprets the dvc.lock and dvc.yaml files to create interactive data pipelines. These pipelines are versioned and contain critical information about files, metrics, and data steps.

How to Use DVC with Outpost?

Configuring Outpost Storage with Your Machine

1. Navigate to your repository homepage.
2. Click the remote button and select the Data tab.
3. Choose DVC.
4. Copy the commands provided to set up your local machine with Outpost Storage.

Enter a terminal in your project, paste the commands, and execute them:

1dvc remote add origin s3://dvc
2dvc remote modify origin endpointurl https://outpost.run/<Outpost-user-name>/<repo_name>.s3 or store.outpost.run/user/repo s3:// gs://
3dvc remote modify origin --local access_key_id <Token>
4dvc remote modify origin --local secret_access_key <Token>