Overview
About e-INFRA CZ Onedata
What is Onedata?
Onedata is a global data management platform that lets you store, share, and access research data across distributed storage systems — regardless of where the physical storage is located. It presents data from multiple storage providers as a single, unified file system, making it easy to work with large datasets across institutions and borders.
In simple terms: Think of Onedata as a smart cloud storage system designed specifically for research — one that connects storage from multiple locations and lets you access it all from one place.
When to Use Onedata & Its Advantages
When is Onedata a Good Fit?
Onedata is well-suited for:
- Storing and sharing large research datasets that exceed the capacity of typical cloud storage (e.g., Google Drive, Dropbox)
- Collaborating with colleagues across different institutions or countries
- Archiving research outputs that need to be accessible long-term
- Processing data on HPC (High-Performance Computing) clusters — Onedata can be mounted directly as a file system
- Publishing datasets so others can access or cite your research data
Key Advantages
| Feature | Benefit |
|---|---|
| Unified access | Access all your data from one place, no matter where it’s physically stored |
| High capacity | Designed for large-scale scientific datasets |
| Flexible sharing | Share data with specific users, groups, or make it publicly accessible |
| Data locality transparency | See which storage provider holds your data |
| Multiple access methods | Web portal, desktop sync, command-line, REST API, FUSE mount |
| Security & access control | Fine-grained permissions for files and folders |
| Integration with HPC | Can be used directly with computing resources |
Not sure whether to use Onedata or S3? Here is a quick comparison.
Differences between the two:
| Onedata | S3 Storage | |
|---|---|---|
| Purpose | Data management platform across a storage federation | Raw data storage |
| Storage type | Multiple backends with a unifying layer over them | Object storage |
| Access methods | POSIX FS, S3 API, REST API, CDMI API | S3 API only |
| Data location | Same data can live in multiple locations | Single data storage cluster |
| Other features | Permissions, groups, roles, replication, inter-site transfer | Tenant/bucket/object model, reliability |
When to use:
| Onedata |
|---|
| Unified interface across multiple storages. |
| Multiple access methods. |
| Integration of already existing storage. |
| Distributed / shared datasets, operations on data from different locations. |
| S3 Storage |
|---|
| Backup, archiving. |
| Fast access, big data. |
| No need for automatic migration, replication, sharing. |
| S3 API access is sufficient. |
Official documentation at Onedata docs.
Service guarantees
Onedata is a data management platform, not a storage. Therefore the system has limited guarantees when it comes to data loss and consistency.
The overall policy is based on Terms and Conditions of CESNET Data Storage Services.
e-INFRA CZ Onedata puts priority on keeping the service operational. We do regular backups of runtime data of all components and are able to restore the system to the last functional state in case of emergency. The frequency of the backup aims to minimize potential problems with restoring. However, some inconsistencies in user data might still occur. See also the Data availability details section.
