EGI (e-Infrastructure)

EGI (e-Infrastructure)
Name	EGI (e-Infrastructure)
Formation	2010
Type	Research infrastructure
Headquarters	Amsterdam
Region served	Europe, global
Parent organization	EGI Foundation

EGI (e-Infrastructure) is a distributed computing and data infrastructure that provides large-scale high-performance computing and grid computing services for research communities across Europe and globally, supporting multidisciplinary science in fields such as astronomy, bioinformatics, climate science, and particle physics. The infrastructure connects national and institutional computing centres to deliver federated compute, storage, and data management, enabling collaborations among projects like CERN experiments, European Space Agency, and international consortia including ELIXIR and Human Brain Project.

Overview

EGI provides a pan-European federated platform combining resources from national research and education networks such as GÉANT and national nodes like SURFsara and CESNET, integrating services used by communities working with facilities such as European Synchrotron Radiation Facility, Max Planck Society, European Molecular Biology Laboratory, and Joint European Torus. The infrastructure interoperates with initiatives like PRACE and projects under the Horizon 2020 and Horizon Europe programmes, offering services aligned with standards from organisations including OGF, IEEE, and OpenStack Foundation. EGI enables reproducible research and large-scale workflows used by collaborations related to LHC, Square Kilometre Array, Copernicus Programme, and global initiatives like the Global Earth Observation System of Systems.

History and Development

EGI evolved from early grid efforts such as EDG and EGEE, which supported experiments at CERN and collaborations like ALICE, ATLAS, CMS, and LHCb. The formal transition to a federated e-infrastructure followed strategic reviews from the European Commission and stakeholders including European Research Area bodies and national research councils exemplified by Science Foundation Ireland and Deutsches Forschungszentrum für Künstliche Intelligenz. Key milestones include formation of the EGI Council, establishment of the EGI Foundation in Amsterdam, and partnerships with projects under FP7 and later Horizon 2020. The roadmap intersected with initiatives such as OpenAIRE, EOSC and technology providers like Red Hat, Canonical, Intel, and vendors represented at events like Supercomputing Conference.

Architecture and Services

EGI’s architecture is federated, combining computing elements such as HTCondor pools, Slurm clusters, and OpenStack clouds, with storage solutions from Ceph, dCache, and object stores inspired by Amazon S3. Core services include identity and access management via federations like eduGAIN and token services using standards from OAuth and SAML, data transfer tools akin to GridFTP and Globus, and monitoring stacks integrating Nagios and Prometheus. The service catalogue supports virtual research environments, container orchestration with Kubernetes, and workflow engines comparable to Apache Airflow and Nextflow, enabling scientists affiliated with institutions like University of Oxford, ETH Zurich, Sorbonne University, and Karolinska Institutet to run distributed analyses.

Governance and Organisation

EGI is coordinated by the EGI Foundation, governed through a membership and stakeholder framework involving national resource providers such as CESGA, FZJ, and regional e-infrastructure consortia like Reduce, with advisory input from boards and committees paralleling models used by European XFEL and ESFRI projects. Strategic oversight engages funding bodies including European Research Council, national ministries exemplified by Ministry of Education, Culture and Science (Netherlands), and research organisations such as CERN and European Commission directorates. Governance mechanisms align with policies from GDPR for data protection and adhere to quality frameworks similar to ISO certifications.

Federation and Membership

Membership comprises national operators, research institutions, and service providers drawn from countries across European Union, Norway, Switzerland, and partners in South Africa, Brazil, and China, collaborating through federated agreements comparable to EUMETSAT arrangements. The federation model interoperates with infrastructures like National Grid Service (UK), XSEDE, and regional initiatives such as PRAGMA, enabling single sign-on via eduGAIN and resource sharing under service level agreements shaped by organisations like TERENA.

Use Cases and Scientific Impact

EGI supports research in domains including astrophysics projects tied to LOFAR and SKA, genomics consortia like ELIXIR and 1000 Genomes Project, climate modelling used by ECMWF, and computational chemistry collaborations with groups at Max Planck Institute for Chemical Physics of Solids. It has facilitated data-intensive workflows for LIGO analysis, combined analyses across facilities like EURATOM fusion experiments, and supported digital humanities projects at institutions including King's College London and University of Bologna. Outcomes include publications in journals such as Nature, Science, and Physical Review Letters, and contributions to policy reports by IPCC and technical standards in bodies like ETSI.

Funding and Sustainability

Funding stems from a mix of European programmes (Horizon Europe, FP7, Horizon 2020), national research agencies such as UK Research and Innovation, Deutsche Forschungsgemeinschaft, and project-based contracts with organisations like CERN and ESA. Long-term sustainability strategies reference models used by PRACE and EuroHPC, combining membership fees, service income from academic partners including University of Cambridge and École Polytechnique, and competitive grants, while engaging with industry partners such as IBM, Microsoft, and Google for technology collaborations and joint innovation initiatives.

Category:Research infrastructures