#	Pagina
attuale pagina	/open-h2020/projects/220798/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - ICEI (Interactive Computing E-Infrastructure for the Human Brain Project)

Teaser

Summary

ICEI (Interactive Computing e-infrastructure for the Human Brain Project) is an EU-funded project under the Framework Partnership Agreement of the Human Brain Project (HBP). Five leading European Supercomputing Centres are working together to develop a set of HPC and Cloud services that will be federated to build the foundation of the Fenix Infrastructure. The centres (BSC, CEA, CINECA, ETHZ-CSCS and JUELICH-JSC) committed to perform a coordinated procurement of equipment, licences for software components and R&D services to realize elements of this e-infrastructure. Fenix infrastructure services include elastic access to scalable and interactive compute resources, VM services and a federated data infrastructure. The distinguishing characteristic of this e-infrastructure is that data repositories and scalable supercomputing systems are in close proximity and well integrated.
The Fenix Infrastructure will deliver federated compute and data services to European researchers by aggregating capacity from multiple resource providers and enabling access from existing community platforms, like for the HBP. User access is granted via a peer-review based process. The HBP is the initial prime and lead user community, guiding the infrastructure development in a use-case driven co-design approach. Access for other science and engineering communities is provided through PRACE.
ICEI aims to support science communities that develop, deploy, and operate domain-specific platform services by providing computing and storage resources as a basis for community-specific platforms and services, e.g. HBP Collaboratory (https://wiki.humanbrainproject.eu) and Materials Cloud platform (https://www.materialscloud.org).

Work performed

The project first focussed on defining the technical requirements for the ICEI Infrastructure that provide a basis for the preparation of the coordinated procurement of equipment, licences for software components, and R&D services. In order to specify common technical requirements, information was collected from different sources, including a Public Information Event (PIE), and Request For Information (RFI) meetings. Science and use cases have been collected from the HBP. They are continuously analysed with respect to their implementation progresses towards meeting their specific e-infrastructure needs and requirements. Specifications and concepts for the procurement procedures were developed based on the architecture specification, which was defined so that the constraints of each data centre were taken into account. This concerned, in particular, the AAI strategy, which has been endorsed by all involved sites. The documents for all procurements have been prepared, some have been published. Two procurements have been completed and contracts awarded. While most of the infrastructure services will become available in 2020, there are already compute and storage resources available at ETHZ-CSCS. The project defined and set-up a peer-review based process to allocate resources to users from HBP and to users applying for access through PRACE calls. Currently, there are more than 10 projects using resources provided by the ICEI project. The project has addressed various ethics questions and, in particular, has setup procedures that enable the processing of a certain type of personal data, which is called pseudonymised data. Since the start of the project various communication and dissemination channels have been established, notably a web page used for providing information for the general audience, users of the infrastructure as well as suppliers of IT equipment and R&D services.

Final results

The ICEI project has laid the basis for expanding the e-infrastructure services portfolio beyond what is offered by European supercomputing centres today. While the focus of these centres was so far on providing scalable computing capabilities, ICEI expands this by providing the opportunity to use tightly integrated servers interactively, e.g. for data pre- and post-processing and visualisation. Users will also be able to deploy services in virtual machines. This will, for instance, allow deploying science domain-specific gateways, i.e. web-based interfaces for providing a variety of IT resources and services to researchers of a particular science domain. Furthermore, data repositories with different characteristics are being provided, which allow sharing data, hold extreme-scale data volumes and/or have high-performance access to data. ICEI is not only progressing beyond the state-of-the-art by expanding the service portfolio but also by federating these services. Users can therefore not only flexibly combine services at different locations but can also more easily collaborate within their respective science communities.
Enabling new forms of collaborative research using modern computational and data science methods will allow advancing different science areas with high societal impact. Examples are neuroscience research resulting in new approaches to treat brain diseases, materials science resulting in the discovery and design of new materials that are key for industrial innovation and competitiveness.