Get started
Fusion is a distributed virtual file system optimized for cloud-native data pipelines. It provides fast, efficient access to cloud object storage (AWS S3, Azure Blob Storage, Google Cloud Storage) by allowing your pipeline tasks to work with cloud data as if it were a local POSIX file system, eliminating the need for data staging and transfer.
Use Fusion directly in Seqera Platform compute environments, or add Fusion to your Nextflow pipeline configuration.
Seqera Platform
Fusion is available in the following Seqera Platform compute environments:
See the compute environment documentation for your specific cloud provider above for detailed configuration instructions, required permissions, and optimal compute and storage recommendations.
Nextflow
Fusion integrates seamlessly with Nextflow and requires no installation or pipeline code changes.
This section describes how to enable Fusion in your Nextflow pipelines. It includes:
- Creating a Platform access token
- Configuring Fusion in your pipeline
- Running your Nextflow pipeline with Fusion
You will need the following to get started:
- Nextflow 22.10.0 or later installed locally
- A Seqera account
- A configured container runtime or container computing service (e.g., Kubernetes, AWS Batch, or Google Cloud Batch)
Create your Platform access token
A Platform access token is your personal authentication key that enables access to Seqera Platform services.
To create a Platform access token:
- Log in to Seqera
- From your personal workspace, go to the user menu and select Settings > Your tokens
- Select Add token
- Enter a unique name for your token, then select Add
- Copy and store your token securely
The access token is displayed only once. Save the token value before you close the Personal Access Token window.
Configure Fusion
To enable Fusion, add the following to your nextflow.config file:
fusion.enabled = true
wave.enabled = true
tower.accessToken = '<TOWER_ACCESS_TOKEN>'
Replace <TOWER_ACCESS_TOKEN> with your Platform access token.
Run your pipeline
Run your pipeline as you normally would. For example:
nextflow run main.nf
Fusion will automatically provision and mount your pipeline storage.