High-performance computing
EBRAINS provides computing and storage services at leading European supercomputing and data centres, which are part of the Fenix research infrastructure. Fenix is realised through the Interactive Computing E-Infrastructure (ICEI) project. These services can be requested and used separately or in combination, according to your needs. They include Scalable and Interactive Computing Services, e.g. for large-scale simulations and analyses, Virtual Machine Services for running customised environments, and Archival and Active Data Repositories for storing, sharing and accessing data.
- Scalable Computing Services that can, e.g., be used for large-scale simulations or heavy machine-learning workloads.
- Interactive Computing Services that allow using high-end servers, e.g., for interactive data analysis.
- Virtual Machine Services to run your customised environment in a cloud environment.
- Archival Data Repositories for storing data in preparation for sharing through EBRAINS Data and Knowledge Services.
- Active Data Repositories for fast access to private copies of your data assets from the supercomputer.
Accessing services
Europe-based neuroscientists are eligible to apply for access to Fenix services, in particular, whose research falls into the following subjects:
- Analysis of brain data leading to understanding brain function
- Building of multi-scale scaffold theory and models for the brain
- Simulation of brain function
- Brain-inspired computing, data analytics and robotics
Details related to eligibility criteria are available in the call texts. Although the applying Principle Investigator needs to be based in Europe, the project members can also be working outside of Europe
The project results are expected to lead to a significant European added value and major impact at the international level.
Proposals can be submitted at any time. Applicants should target allocation of resources that are at most 12 months. Renewal of applications is possible if the project exceeds this period.
Proposals will be processed by the EBRAINS Infrastructure Allocation Committee (IAC) which manages the review of technical viability (supported by technical teams at the Fenix sites) and scientific excellence (possibly with experts in the corresponding scientific field).
Important: Due to the end of the HBP SGA3 and ICEI projects, the allocation process for HBP/EBRAINS, as described above, is currently closed. Please visit the Fenix website in order to check the availability of ICEI resources at the different Fenix sites: https://fenix-ri.eu/infrastructure/resources/available-resources. For more information regarding the status of your project, please send an email to icei-coord@fz-juelich.de.
The e-infrastructure services
The distinguishing characteristic of the Fenix Infrastructure and its services is that data repositories and scalable supercomputing systems will be in close proximity and well integrated.
Below is a short description of all available Fenix services for research communities and users. Click on each title for more details on the specific services.
Interactive Computing Services
Quick access to single compute servers to analyse and visualise data interactively or to connect to running simulations, which are using scalable computing services.
Scalable Computing Services
Massively parallel HPC systems that are suitable for highly parallel brain simulations or high-throughput data analysis tasks.
Virtual Machine Services
Service for deploying virtual machines (VMs) in a stable and controlled environment is suitable for deploying platform services like the EBRAINS Collaboratory, image services or neuromorphic computing front-end services.
Active Data Repositories
Site-local data repositories close to computational and/or visualisation resources that are used for storing temporary replicas of data sets. In the near future, they will typically be realised using parallel file systems.
Archival Data Repositories
Federated data storage, optimised for capacity, reliability and availability, is used for long-term storage of large data sets which cannot be easily regenerated. These data stores allow data sharing with other researchers inside and outside EBRAINS.
A list of all available and planned e-infrastructure services can be found on the Fenix website.