Context
The EDITO Data Lake is the data-access component of EDITO. It is composed of two complementary services:
EDITO Data Storage stores data objects in S3-compatible buckets.
EDITO Data Catalogue stores metadata and references to datasets. It can reference data hosted in EDITO storage or on an external platform.
📌 Note: adding a file to EDITO storage does not automatically publish it in the Data Catalogue. Likewise, a dataset referenced in the Catalogue does not have to be physically stored by EDITO.
Understand EDITO storage
EDITO provides permanent S3-compatible object storage (or elastic cloud object storage) for personal work, group projects and public data. You can also connect storage hosted by another S3 provider.
This article explains the available storage options and how they relate to temporary computing storage and the EDITO Data Catalogue.
How S3 object storage differs from a filesystem
EDITO storage is compatible with the AWS S3 API, although the default EDITO storage is not hosted on Amazon Web Services.
S3 object storage uses three main concepts:
A bucket is the main container for your objects.
An object is a file-like item stored in a bucket.
An object key is the complete name of the object inside the bucket, for example
results/2026/output.nc.
The folders displayed in File Explorer are not real directories. They are prefixes in object keys. For example, results/2026/ is the common prefix of all objects whose keys begin with that text.
💡 This has a few practical consequences:
You do not need to create a folder before uploading an object into it.
Moving or renaming a folder usually means copying or moving all objects under a prefix.
Do not start an S3 object key with
/unless a specific tool explicitly requires it.
What storage is available on EDITO?
Personal storage
Every EDITO user has a personal S3 bucket.
It is managed by the user.
Its content is private by default.
The user can make selected files or prefixes publicly accessible.
The default bucket name is usually
oidc-[YOUR_USERNAME].It is normally hosted on the EDITO MinIO endpoint
minio.dive.edito.euin regionwaw3-1.The current default quota is 20 GB of storage.
Use personal storage for files, notebooks, intermediate results and datasets that belong to your own EDITO workspace.
Group project storage
A group project has a shared S3 bucket managed by the project members.
All project members can access the group-project storage.
Project settings, including S3 configurations, are shared with all project members.
The current default quota is 50 GB of storage.
Members can decide which content remains private to the project and which content becomes public.
Use group-project storage for files and datasets that must be available to several members of the same EDITO project.
⚠️ Access to group-project storage is associated with each member's personal MinIO identity. A group project does not have its own personal MinIO token.
Do not attach your unrestricted personal MinIO token to a service shared with the whole group (see Access EDITO storage from code, services and processes).
Additional project storage
Projects that need more capacity can request an additional bucket from the EDITO User Support. This request must be justified.
The additional bucket may be hosted by EDITO's cloud provider, on a CloudFerro S3 endpoint such as s3.waw3-1.cloudferro.com. Depending on the project configuration, it may be available in addition to the default MinIO bucket or used as the main project storage.
You can check the endpoint, bucket or working-directory path and credentials in:
External S3-compatible storage
You can connect an AWS, OVH, CloudFerro or other S3-compatible bucket to EDITO.
An external bucket can be used:
Alongside the default EDITO bucket;
Instead of the default bucket for a particular workflow;
As a public anonymous data source;
As a private bucket using credentials supplied by its provider.
The capacity, retention policy, billing and permissions of an external bucket are managed by its provider or owner, not by the EDITO personal-storage quota (see Connect an external S3 storage to EDITO).
Temporary storage vs permanent storage
The filesystem inside a running service or process is not the same as your permanent S3 storage.
Storage | Persistent? | Typical use |
Service/process filesystem | No | Temporary computation |
Personal EDITO storage | Yes | Personal files/results |
Group project storage | Yes | Shared project files |
Additional/project external storage | Yes | Larger/project-specific datasets |
Public storage | Yes | Public EDITO datasets |
⚠️ A service can be stopped, restarted or removed during platform maintenance. Its working filesystem may therefore be lost. Save source code in Git or in permanent storage, and copy important outputs to S3.
The Open in Jupyter Notebook action in the File Explorer copies selected objects into the Jupyter service's working directory. Changes made there are not automatically written back to S3. Copy the modified files back to permanent storage before stopping the service.
Where to find and configure storage
File Explorer
Use File Explorer to browse the bucket currently selected for EDITO explorers. From there you can upload, download, rename, delete, open or share objects.
Project Settings
Open Project Settings → S3 Configurations to:
Inspect the current bucket configuration;
Connect an external bucket;
Change which S3 configuration is passed to services and processes;
Choose which configured bucket is displayed in File Explorer with Use for EDITO explorers.
⚠️ When you're in Project Settings, pay attention to the Use for EDITO Explorers option. If you have multiple buckets, you'll only see the one with this option enabled in File Explorer.
My Account → Connect to storage
Your personal storage access variables are available in My Account → Connect to storage. This section also lets you copy or download an initialization script in a supported programming language for access from outside a Datalab service:
Services and Processes
When a compatible service or process is launched, its S3 configuration is prefilled from the current project settings. The resulting endpoint and credentials are injected into the runtime as environment variables.
Storage ownership and visibility
Owners and authorized project members can make selected objects or prefixes public. Public access can later be removed.
Making an object public only changes access to the stored data. It does not create Catalogue metadata or make the dataset discoverable in the EDITO Data Catalogue (see Share files and datasets from EDITO storage).
Storage type | Managed by | Default Visibility |
Personal storage | The user | Private |
Group project storage | Project members | Restricted to the project space |
External bucket | Bucket owner/provider | Defined by the external configuration |
Storage quotas
EDITO is public and share resources funded by the European Commission. To avoid abuses, users have usage quotas. The following table summarizes the current quotas for personal and group projects:
Project kind | Storage volume (GB) | CPU vCores | RAM (GB) | Web Disk Storage SSD (GB) | GPU vCores | max. pod count |
Personal project | 20 | 8 | 32 | 50 | 0/1 | 50 |
Group project | 50 | 16 | 64 | 100 | 0/1 | 100 |
📌 Note: “pods” are entities in which services and processes run. For simplicity, one can consider a service or process needs one pod to run.
If your services or processes never launch, face performance issues, or are stopped without an explicit action of yours, the root cause might be due to these restrictions.
Contact EDITO User Support through the chat widget if your work requires a higher quota. Explain the expected volume, purpose, duration and whether the data is personal, project-related or intended for public onboarding.
⚠️ The default resources do not include a GPU (this applies to both personal and project spaces). To obtain one, you must submit an explicit and justified request to support.
Which storage should I use?
I am working alone and need persistent files
→ Use personal EDITO storage.
I am collaborating with an EDITO group/project
→ Use group-project storage.
My project needs much more storage
→ Request additional project storage or request an external S3 bucket.
My organization already has an S3 bucket
→ Connect that external bucket to EDITO.
I only need scratch space during a computation
→ Use the service or process filesystem, then copy important outputs to permanent S3 storage.
I want people outside EDITO to read my data
→ Store it in a bucket you control (personal/project) and configure public read access for the relevant object or prefix.
I want the dataset to be discoverable through EDITO
→ Host the data at an accessible location and register its metadata in the EDITO Data Catalogue.
What's next?
If you have any questions, problems, or suggestions, please feel free to contact us via chat using the widget available at the bottom right of the page.