Skip to content

Credits

Jupyter4NFDI provides access to powerful, flexible computing resources offered by different research institutions across Germany. To ensure that all users can benefit fairly from these shared resources, Jupyter4NFDI introduces a credit-based resource usage system.

This system helps prevent situations where a small number of users consume the majority of the available resources. Instead, credits allow for a balanced and predictable usage model where everyone receives an equitable opportunity to work with large compute environments.

What Are Credits?

Credits represent a user's ability to start and run computing environments (called Flavors) on a Jupyter4NFDI system.
Every user receives an allocation of credits. As long as credits are available, users can continue running sessions on the platform.

The amount of credits you receive may depend on your selected AAI (Authentication and Authorization Infrastructure) (more). Different AAIs may lead to different credit budgets.

How Credit Consumption Works

When you start a session on Jupyter4NFDI, you choose a Flavor, which defines the amount of CPU, memory, storage, or GPUs your environment will have.
Credit consumption is tied directly to this choice:

  • Lightweight Flavors consume fewer credits over time.
  • Large or GPU-enabled Flavors consume credits at a faster rate.
  • Credits are deducted continuously while your session is running.

If your credits reach zero, sessions will be stopped automatically. You will receive new credits every few hours.

Project-Based Credits

Jupyter4NFDI supports collaboration through Projects.
A Project is a shared credit pool that can be used by multiple researchers. This is particularly useful in situations such as:

  • Research teams working on joint analyses
  • Organized workshops
  • Communities using Jupyter4NFDI as shared infrastructure

When you are a member of a Project:

  1. Your resource usage first deducts credits from the Project pool.
  2. Only if the Project credits are exhausted will your personal credits be used.

This ensures collaborative work is not limited by individual user budgets and creates a seamless environment for group-based scenarios.

If your team needs additional credits, you can contact us and request a Project.

Per-System Credit Allocation

Jupyter4NFDI operates across multiple resource providers.
Each provider manages its own credit configuration, meaning:

  • You may see different credit amounts depending on the system you access.
  • Some systems may offer higher or lower credit budgets.
  • Providers can adjust credit allocations and consumption rates over time, depending on load, hardware availability, or policy changes.

Your displayed credit amounts and consumption are therefore system-specific.

Summary

The Jupyter4NFDI Credit System ensures a fair, transparent, and flexible way of distributing computational resources. Users benefit by:

  • Avoiding resource monopolization by heavy users
  • Being able to plan usage based on predictable credit consumption
  • Collaborating through shared Project credit pools
  • Accessing multiple systems with different credit configurations

If you have questions about your credits, Flavors, or how to request a Project, please contact us.