OSCARS Composability & AAI Workshop

CERN, 18th- 19th September 18th, 2025

​

Welcome to CERN!

• Many thanks for making the trip to Geneva
• This meeting has been largely organised by Giovanni - many thanks to him!
• Two tour has been organised today (16:45) and tomorrow (08:00)
• Make sure to visit the (relatively) new Science Gateway if you get a chance

Collaborative Notes Document

​

Please feel free to contribute to our shared notes document: https://tinyurl.com/OSCARSCERN

Workshop Agenda - Thursday 18th September

1. Welcome & Registration (09:00)
2. Introduction (09:30)
3. Summary of the AARC-TREE Compendium workshop (10:00)
4. Introducing the OSCARS Composability Scenarios + Discussion (10:30)
1. ENVRI - Joaquín López Lérida
2. ESCAPE - Giovanni Guerrieri, Marion Pierre, Leo Chazallet
5. Coffee Break (11:15)
6. Introducing the OSCARS Composability Scenarios + Discussion (11:30)
• LSRI - Romain David
• PANOSC - Paul Millar & Melanie Nentwich
• SSHOC - Sally Chambers, Laure Barbot, Matej Ďurčo, Michael Kurzmeier
7. Lunch - CERN Restaurant (12:30)

​

OSCARS Composability & AAI Workshop

CERN, 18th- 19th September 18th, 2025

​

Collaborative Notes Document

​

Please feel free to contribute to our shared notes document: https://tinyurl.com/OSCARSCERN

Workshop Agenda - Thursday 18th September

• Welcome & Registration (09:00)
• Introduction (09:30)
• Summary of the AARC-TREE Compendium workshop (10:00)
• Introducing the OSCARS Composability Scenarios + Discussion (10:30)
• ENVRI - Joaquín López Lérida
• ESCAPE - Giovanni Guerrieri, Marion Pierre, Leo Chazallet
• Coffee Break (11:15)
• Introducing the OSCARS Composability Scenarios + Discussion (11:30)
• LSRI - Romain David
• PANOSC - Paul Millar & Melanie Nentwich
• SSHOC - Sally Chambers, Laure Barbot, Matej Ďurčo, Michael Kurzmeier
• Lunch - CERN Restaurant (12:30)

​

Workshop Agenda - Thursday 18th September

• Introducing the EOSC Federation Buildup Phase, Bob Jones, EOSC Association /CERN (13:30)

​

• Introduction to the EOSC-EU node Spiros Athanasiou (13:45)

​

• Presentation of the CERN EOSC node (ESCAPE) Giovanni Guerrieri and colleagues (14:15)

​

• Panel Discussion & Q&A (14:45)

​

• Coffee break (15:00)

​

​

Workshop Agenda - Thursday 18th September

13. CESSDA Services Integration (SSHOC), Alen Vodopijevec (15:15)

​

• Resource discovery: Challenges & Solutions, Moderated by Matej Ďurčo (15:45)
1. Presenting the problem: fragmented access to data - Anca Hienola
2. LifeWatch Demo: https://search.lifewatch.eu/ - Joaquín López Lérida
3. CLARIN - Centre Registry - VLO Curation effort - Alex König
4. Discussion - How can we improve our domain-specific / cluster data discovery ?

​

• CERN facilities visit (16:45 - 18:00)

​

• Dinner (20:00)

​

How to get to R1 without running a marathon

​

• Directions from the CERN Maps app

Today’s Visits

We were here yesterday

We are here now

Tomorrow’s Visit

ATLAS Experiment visit on Friday, September 19th at 8:00AM

​

We meet in front of building 160

Working Dinner

• We had fondue yesterday, let’s take a break

​

• Reservation for 20:00 for 30 people at Luigia (name Giovanni)

​

• Rue Adrien-Lachenal 24a, 1207 Genève

​

• Nicest walk = Catch tram 18 down to Bel Air and walk through the old town

https://science-clusters.eu/

https://oscars-project.eu/

Interdisciplinary Collaboration

OSCARS Cascading Grants Calls

https://oscars-project.eu/projects

AARC-TREE Project

18

https://aarc-community.org/aarc-tree-project/

https://aarc-community.org

Introduction to AARC TREE

• EC funded project to develop and promote best practices in Federated Authentication and Authorisation for Research Infrastructures
• Part of HORIZON-INFRA-2023-DEV-01-05 (CSA): Preparation of common strategies for future development of RI technologies and services within broad RI communities
• 3rd iteration of the AARC Project
• Main output is guidelines that are being compiled into a “Compendium of Best Practice”
• Close links with EOSC AAI
• March 2024 - February 2026
• Coordinator: Licia Florio - licia at nordu.net
• Wiki: https://wiki.geant.org/spaces/AARC/pages/738885704/AARC+TREE+Project
• Website: https://aarc-community.org/architecture

19

https://aarc-community.org

AARC-TREE Compendium and Recommendations

20

Aim: To produce a compendium of AARC best practices and deliver recommendations for a common long-term strategy for AAI services in pan-European Research Infrastructures in Europe

​

• Builds on RI Use Cases based on stakeholder interviews
• Takes into account the Changing landscape, e.g. establishment of the EOSC EU Node, EOSC Federation and EOSC ‘Candidate’ Nodes etc.

https://aarc-community.org

AARC Compendium Workshop Summary

What did we do?

• Introduced the first draft of the AARC Compendium (input welcome any time!)
• Gathered feedback from the room
• Summary of the AARC Use Cases survey - what are the current AAI challenges of research communities?
• AAI Challenges from 4 specific communities: SSH Open Marketplace, EISCAT, Sea Data Net, Life Sciences
• Update on OIDC Federation (necessary for EOSC and other AAI-to-AAI trust use cases)
• Update of the AARC Compliance Validation work, including a demo of a self-service online validation tool
• Had a very nice Fondue!

21

Apologies to the online attendees for the atrocious timekeeping in the room!

https://aarc-community.org

AARC Compendium Workshop Summary

Main takeaways

• Areas of focus for Compendium
• Target/split information for specific user groups (e.g. funders, managers, technical implementers)
• Improve glossary and ensure consistency
• Try to be more helpful in guiding communities to select hosted vs self-hosted
• Listing services/software is very useful but more information is needed
• We should clarify why using large commercial AAIs are usually not the right choice for research communities
• A “consultancy” type help will always be necessary - can we leverage other groups? Can we use AI?

​

This was incredibly helpful for improving the Compendium and, based on feedback from the room and the levels of engagement, we believe it was a valuable experience for participants. Thank you for your energy!

22

https://aarc-community.org

OSCARS Consolidation Workshop

https://oscars-project.eu/news/oscars-consolidation-and-terminology-workshop-overview

Consolidation and Terminology Workshop, 30th September - 1st October 2024, DESY, Hamburg.

Research Data Lifecycle

https://rdmkit.elixir-europe.org/data_life_cycle

https://www.gida-global.org/care

https://www.go-fair.org/fair-principles/

Towards Reproducible Research Workflows

Creating reproducible research workflows based on Open Science practices involves the steps of:

• FAIRification of the data
• the interoperability of tools and services needed to analyse that data and
• the sustainable publication of the research outputs, such as publications, datasets, software etc.

From Consolidation to Composability

Composability Action Plan: provide an initial outline of the steps to implement the research scenarios described in D2.1 Clusters Services and Data Sources Portfolios as composable workflows using Open Science practices.

ENVRI Hub & OSCARS Composability Scenario

Joaquin LOPEZ LERIDA

LifeWatch ERIC / ENVRI

Composability & AAI workshop

Sept 2025

Context of OSCARS

28

OSCARS Composability & AAI workshop

17 – 19 September 2025

Joaquin Lopez – LifeWatch ERIC

OSCARS: Open Science Clusters' Action for Research & Society

​

Goal: Enable composability across research infrastructures

​

What is ENVRI Hub:

• Integrated access portal for ENVRI Research Infrastructures
• Provides access to data, services, and computing resources
• Acts as a gateway for environmental research interoperability

https://envrihub.vm.fedcloud.eu/

Role of ENVRI Hub in OSCARS

29

Characteristics:

Supports composability of environmental research services

• Service orchestration via EOSC/FedCloud APIs
• Reuse of standardized workflows and data services

​

Facilitates multidisciplinary experiments

• Cross-domain data discovery and integration (e.g., biodiversity + climate + marine)
• Semantic alignment of metadata using ontologies and FAIR principles

​

Acts as integrator across biodiversity, climate, ecosystems

• Federated Authentication & Authorization (AAI)
• Connection to catalogues, data repositories, and computing backends
• Enables interoperability among LifeWatch, ICOS, eLTER, Euro-Argo and others

​

Technical Foundation

• FAIR Data Services (catalogues, metadata schemas, semantic brokers)
• Workflow engines and Jupyter-based environmentsInteroperability with EOSC Core services (storage, compute, identity)

​

OSCARS Composability & AAI workshop

17 – 19 September 2025

Joaquin Lopez – LifeWatch ERIC

Chosen Composability Scenario

• Semantic integration of heterogeneous environmental datasets
• Use of ontologies and FAIR vocabularies for alignment
• Cross-domain modelling of biodiversity and ecosystem dynamics
• Linking biodiversity data with climate and ecosystem drivers
• Workflow composition through ENVRI Hub services and EOSC/FedCloud
• Demonstrates interoperability and reusable building blocks across RIs

Technical Components

​

• Federated Authentication (AAI)
• FAIR Data services: catalogues, metadata, ontologies
• Compute resources via EOSC and FedCloud

OSCARS Composability & AAI workshop

17 – 19 September 2025

Joaquin Lopez – LifeWatch ERIC

30

Practical Examples

31

OSCARS Composability & AAI workshop

17 – 19 September 2025

Joaquin Lopez – LifeWatch ERIC

Pilot case:

• integration of MARBEFES, LifeWatch search engine, federated datasets
• Researcher workflow:
• Discover data and services
• Compose workflows
• Run analysis on cloud resources

Challenges and next steps:

- Data heterogeneity

- Semantic interoperability

- User adoption

​

Next Steps:

- Enrich scenarios

- Enhance tools and services

- Collect feedback from clusters

OSCARS Composability & AAI Workshop

CERN, 18th- 19th September 18th, 2025

​

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Romain David (ERINHA - LS-RI Cluster)

Workshop Agenda - Thursday 18th September

1. Welcome & Registration (09:00)
2. Introduction (09:30)
3. Summary of the AARC-TREE Compendium workshop (10:00)
4. Introducing the OSCARS Composability Scenarios + Discussion (10:30)
1. ENVRI - Joaquín López Lérida
2. ESCAPE - Giovanni Guerrieri, Marion Pierre, Leo Chazallet
5. Coffee Break (11:15)
6. Introducing the OSCARS Composability Scenarios + Discussion (11:30)
• LSRI - Romain David
• PANOSC - Paul Millar & Melanie Nentwich
• SSHOC - Sally Chambers, Laure Barbot, Matej Ďurčo, Michael Kurzmeier
7. Lunch - CERN Restaurant (12:30)

​

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Composability = ability to assemble modular services related to data and workflows handling across Life Science Research Infrastructures (LS-RIs).�

LS-RIs Scenarios explore how CCs can:

• Combine domain-specific and cross-disciplinary tools.
• Enable interoperability and reuse of services.
• Support scalable and sustainable FAIR practices.

Goal: flexible “plug-and-play” approaches where LS-RIs co-develop and share components, fostering innovation and efficiency.

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

First step for Composability

​

What are the possible �components in LS-RI Cluster?

​

Step 1 �cataloguing Research services and resources provided by LS-RIs categorised into data & databases, technologies and facilities, computational tools, expertise & support, and materials.

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Ten Key Challenges for CCs in LS-RIs

Foundational

1. Definition & Consensus – lack of shared vision on CCs.
2. Awareness & Community Engagement – low visibility and adoption.

Policy� 3. Governance Complexity – unclear or burdensome governance structures.� 4. Incentives for FAIR & Open Science – limited rewards for good practices.� 5. Funding & Policy Alignment – inconsistent support and EOSC integration.� 6. Sustainability & Resource Allocation – need for stable human/financial resources.

Technical� 7. Interoperability & Standardisation – fragmented tools, lack of harmonised standards.� 8. Data Stewardship & Quality Control – gaps in expertise and practices.� 9. Heterogeneous Expertise & Skills Development – uneven training and skills across RIs.� 10. Technical Infrastructure Architecture – need for federated, FAIR-compliant cloud and compute systems

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Three Scenarios for LS-RI Competence Centres

1. Federated Model

• Distributed roles across multiple CCs.
• Preserves local expertise & flexibility.
• Needs strong coordination to avoid fragmentation.

2. Centralised Model

• One central CC entity leads all activities.
• Enables standardisation & efficient governance.
• Less flexible, risk of limited community buy-in.

3. Hybrid Hub-and-Spoke Model

• Central hub ensures coherence; domain-specific CCs act as spokes.
• Balances alignment and adaptability.
• Emerging as the most sustainable and scalable option

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Scenarios vs. Challenges

• Federated Model
• Pros: maximises local expertise, flexibility, diversity of solutions.
• Cons: risk of fragmentation, higher coordination costs, uneven standards.�
• Centralised Model
• Pros: strong governance, clear accountability, easier policy alignment.
• Cons: less flexibility, may reduce community engagement, risk of dominance.�
• Hybrid Hub-and-Spoke Model
• Pros: combines central coherence with distributed expertise.
• Balances sustainability, interoperability, and inclusiveness.
• Viewed as the most scalable and community-supported approach

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

CCs Implementation in LS-RI Cluster�Taking in account all challenges

Phased approach:

• Start from existing RI services and practices.
• Build gradually through pilots and shared use cases.

Governance & Coordination:

• Balance central EOSC alignment with RI-specific autonomy.
• Hybrid hub-and-spoke model emerging as the most feasible.

Operational focus:

• Support composability of tools, workflows, and data services.
• Provide registries, training, and stewardship support.

Sustainability:

Requires long-term funding, incentives for FAIR practices, and community engagement .

Scenarios for Competence Centres (CCs) in the LS-RI Cluster

Competence Centres (CCs) are transformative opportunities for LS-RIs to strengthen FAIR and Open Science.

They can increase exchange of good practices, promote interdisciplinarity, and foster synergies across domains.

Success depends on:

• Clear definitions, roles, and responsibilities aligned with EOSC.
• Strong community engagement and recognition systems.
• Streamlined governance and cross-cluster interoperability.
• Robust data quality frameworks and incentives for FAIR compliance.

With coordinated strategies, CCs can become sustainable hubs within EOSC, enabling secondary data use, interoperability, and innovation.

Reproducible Research Workflows

Humans-in-the-loop

Automatic Text Recognition (HTR, OCR)

​

Digital (Text) Analysis Workflows

​

• Named Entity Extraction
• Social Network Analysis
• Sentiment & Emotion Mining

​

OSCARS Composability Workshop

SSHOC Scenario 1: Digitised historical newspapers

Objective�To create an efficient workflow for processing, analysing, and visualising digitised historical newspapers. This scenario will assist researchers in handling large corpora of historical texts, focusing on tasks such as text digitisation, Optical Character Recognition (OCR) improvement, metadata extraction, and historical linguistic analysis.

Workflow Steps

1. Data Collection
2. OCR and Text Enhancement
3. Metadata Enrichment and Annotation
4. Text Mining and Qualitative Analysis
5. Collaboration and Sharing

​

https://rdmkit.elixir-europe.org/data_life_cycle

Degrees of freedom - many moving parts

SSHOC scenario 1

Lot of variation along multiple dimensions possible:

• Type of input data - newspaper, books, manuscripts, hand-written, …�+ text-specific variations like language and script, genre or time of creation
• Source and format of data - accessibility, interoperability
• Methods / Tools desired / available

Actors and their goals

SSHOC scenario 1

• CHI professionals as data providers �who have large collections of data and want to process it on large scale
• Researchers as consumers �who want to find and work with existing data - ideally processed, enriched
• Researchers as data creators �who either bring their own data or create derived datasets based on existing data (e.g. through manual annotation)

Collaborative Notes Document

​

Please feel free to contribute to our shared notes document: https://tinyurl.com/OSCARSCERN

(FAIR) Challenges

SSHOC scenario 1

• Scattered landscape (Where to look for relevant data?)
• Heterogeneity of datasets
• Extreme variability of data (language, script, genre, time)
• Quality of metadata – bad coverage, inconsistent vocabularies
• Varying quality of data – bad scans
• Inconsistent access to data – difficulty to get from a PID to metadata and to actual data
• Unclear licensing / rights statements
• Incompatibility between data and tools
• Tools not available or not fine-tuned for many types of data

Workshop Agenda - Friday 19th September

1. CERN Facilities Visit (08:00)
2. Introduction and Welcome to the final day (09:15)
1. Summary of the previous day
2. Goal of the session
3. Comparison of Actionable Workflows (09:30)
• Galaxy & workflowHub (Björn Grüning, Freiburg)
• REANA (Tibor Simko, CERN)
• EWOKS (Loïc Huder, ESRF)
• DIRAC (Andre Sailer, CERN
4. Coffee break (11:30)
5. Composability Scenarios with Workflows - ENVRI Workflow Engine, Joaquín López Lérida (12:00)

​

Workshop Agenda - Friday 19th September

6. Composability Scenarios - ideally using Workflows - SSHOC - Digitised Historical Newspapers Matej Ďurčo, Laure Barbot, Michael Kurzmeier, Sally Chambers (12:30)
7. Lunch Break (CERN Restaurant) (13:00)
8. OSCARS Composability Scenarios: Take-aways, actions & next steps (14:00)
1. ENVRI - Anca Hienola, Joaquín López Lérida
2. ESCAPE - Giovanni Guerrieri, Marion Pierre, Leo Chazallet
3. LSRI - Romain David
4. PANOSC - Paul Millar & Melanie Nentwich
5. SSHOC - Sally Chambers, Laure Barbot, Matej Ďurčo, Michael Kurzmeier
9. WP2 Next Steps (15:00)
• MS8: Composability Progress Report (December 2025)
• MS9: Initial Engagement Concept (December 2025)
• D2.2: Service Composability Report (June 2026)

​

​

OSCARS Composability Workshop

With many thanks to Giovanni Guerrieri, CERN

Comparison of Actionable Workflows

​

​

https://indico.cern.ch/event/1527699/

https://usegalaxy.eu/

https://reanahub.io/

https://ewoks.esrf.fr/en/

https://dirac.readthedocs.io/

OSCARS Composability Workshop

With many thanks to Giovanni Guerrieri, CERN

Comparison of Actionable Workflows

​

1. System Overview & Philosophy: What is your workflow engine’s core design philosophy?
2. Workflow Definition & User Experience: How do users create and manage workflows?
3. Execution & Scalability: Where and how do workflows run?
4. Data Management & Integration: How does your system handle data and integrate with other tools?
5. Reproducibility & Sharing: How do you ensure reproducible science?

https://indico.cern.ch/event/1527699/

ENVRI Hub & OSCARS Composability Scenario

ENVRI Workflow Engine

Joaquin LOPEZ LERIDA

LifeWatch ERIC / ENVRI

Composability & AAI workshop

Sept 2025

Services and “service components” by category in My LifeWatch [my.lifewatch.eu]

Selecting category

Mouse over elements

Requesting metadata

Metadata available from Metadata Catalogue Service [metadatacatalogue.lifewatch.eu]

Opening the service / launching the service component

Service component launched inside the workflow engine

Technical details about workflow engine

• Workflows are divided into workflow service components. Each component consists of a Docker image and a standardized annotation file that defines the component’s parameters, inputs, and outputs.
• A library of available components is hosted in GitLab (soon to include a public registry and a GitHub mirror), containing their metadata and pre-built images. Workflow metadata is also available through the LifeWatch ERIC Metadata Catalogue, in line with FAIR standards for service publication.
• User data is hosted in an S3-based cloud (MinIO), secured with controlled access and presented through a simple visual interface integrated into MyLifeWatch. Alternatively, user data can be stored in the LifeBlock IPFS system, providing a foundation for workflow certification and decentralized dissemination.
• Workflows can be executed by users either through ready-made templates or by composing them from components in the Workflow Studio.
• The workflow execution system is distributed across multiple nodes, deployed on isolated servers in different data centers as well as in a multi-datacenter Kubernetes environment.
• The proposed architecture is heterogeneous, allowing each workflow to run on the server that best matches its requirements. This includes:
• workflows requiring high-throughput access to specific databases (such as Kraken2 PlusPFP),
• workflows requiring specialized resources (e.g., GPU), or
• workflows with heavy computational demands.

​

IPFS System and workflow interaction

Scientifically-Validated Workflows

Crustaceans (01/23) - Selection

Crustaceans (02/23) – Getting info

Crustaceans (03/23) – More info

Crustaceans (04/23) – Metadata Catal.

Crustaceans (05/23) – Instantiate Exec.

Crustaceans (06/23) - Review

Crustaceans (07/23) – More Review

Crustaceans (08/23) - Start

Crustaceans (09/23) – Give a name

Crustaceans (10/23) – First parameter

Crustaceans (11/23) – Second parameter

Crustaceans (12/23) -

Crustaceans (13/23) – Save and launch

Crustaceans (14/23) – Explore execution

Crustaceans (15/23) – Execution status

Crustaceans (16/23) – All workflows

Crustaceans (17/23) – Any time later

Crustaceans (18/23) – Workflow status

Crustaceans (19/23) – Exec. status

Crustaceans (20/23) – Parameters & outputs

Crustaceans (21/23) – Output & log

Crustaceans (22/23) - Download

Crustaceans (23/23) - Content

Restarting at 14:20?

Takeaways

Takeaways and next steps

​

• PaNOSC:
• Very positive feedback of Björn to fsspec
• Potential collaboration on D3A with Galaxy
• Exploring further applications of D3A, e.g. jupyter notebooks
• Revitalizing communication

​

In a general way, the session showed that many research infrastructures share the same AAI challenges – from federation complexity to sustainability. Composability isn’t about adding more building blocks, but about making them reusable and interoperable across domains.

ENVRI next steps

Takeaways and next steps

• ENVRI
• Galaxy - bring the environment, climate and ecology in the ENVRI node somehow :)
• Advancing composability across RIs—the dashboard on the state of the environment being a strong case in point. Others could be invented. Develop a cross-domain “State of the Environment” dashboard, integrating biodiversity, climate, and marine data as a strong showcase of composability.
• Several solutions were presented here to help ENVRI to tackle data fragmentation—a particularly tough one
• Implement semantic harmonisation services (ontologies, FAIR vocabularies) to reduce data fragmentation and enable reusable workflows across RIs.
• Promote generic, reusable workflow components (standardised, containerised services) that can be shared across RIs, lowering duplication and benefiting the whole research community through easier workflow composition.

LS-RI next steps

Takeaways and next steps

​

Reusability of tools and workflows: Methods developed in one cluster can be applied across others and/or different communities.

• Example: Galaxy is widely adopted across disciplines, with users already federated through ELIXIR.
• Zenodo interoperability is another good example

Advancing composability across RIs:

• Composability is intrinsic to Life Sciences research, especially with the rise of multimodal approaches.
• Key question: how can this be practically connected to Competence Centres (CCs)?
• Next step: proposal for an inter-cluster/project CC paper.

Shared challenges with ENVRI:

• Data fragmentation is a common issue.
• Semantic harmonisation and governance for semantic are key enablers for data quality / quality of DM
• Life Science RI data often need to be combined with ENVRI data to tackle societal challenges (e.g. agriculture, pandemics).

ADOPTABILITY Addressing this requires an integrative approach and stronger assistance—not just support.�Open consideration: could tool-centred CCs be part of the solution?

• We (Björn) will promote more CVMFS and fsspec :) -> that could be a good CC case study?

SSHOC next steps

• Scenario 1 - Digitised Historical Newspapers => try to run on Galaxy
• Create MP workflow for Nina’s use case
• Access to data (from Galaxy)
• Integrating (or using already integrated) tools
• Where to store the result data?
• Engagement: Workflow validation with Nina & colleagues; investigating connection with PressMint (CLARIN), Impresso, NewsEye etc.
• Would it be useful to use SnakeMake, CWL or others for our workflows?
• Could we connect actionable workflows to narrative workflows in the SSH Open Marketplace? (add a ‘Run in Galaxy’ button)
• Scenario 2 - Data anonymisation workflow
• To clarify what direction we would like to go in, e.g. Anonymisation / Pseudonymisation. Reassess the current workflow proposal and modify it to match existing capabilities - including available datasets, data repositories, tools and services - with respect to available resources for utilization of Authentication and Authorization Infrastructure.
• To evaluate Galaxy as a workflow management platform
• Discoverability of ‘data’ resources, read: Registration of EOSC Research Product Catalogues in the EOSC EU Node (3.0). Zenodo. https://doi.org/10.5281/zenodo.15516020

​

ESCAPE next steps

• ESCAPE
• Organize services catalog in a ENVRIHUB-like configuration; we liked that :)
• Extensive documentation and training for people interested in VRE workflows / analyses
• Finalize the Zenodo extension (42% there)
• Working on Rucio and Indigo IAM (AAI) to improve the composability (CTAO scenario) - contacts from AARC
• Overcommitment objective: connect the VRE to an HPC and then use Galaxy or Dirac or Ewoks or REANA with it.
• Overcommitment objective 2: common VRE Helm Charts

​

PaNOSC next steps

Takeaways and next steps

• PaNOSC Scenario 1 – Direct DOI Data Access
• Extension of metalink approach to ESRF
• Adding support for D3A at ESRF (open) data portal
• Adding globus extension
• Further development of the fsspec proof-of-concept → proper integration to fsspec

​

• PaNOSC Scenario 2 – PaNET and WayForLight
• Improve API of WFL
• Optional: Including sample environment information to database (first: create metadata scheme)

​

• Following some more composability ideas
• Ewoks integration in Galaxy?
• Galaxy features are in line with Ewoks next steps
• May be challenging to start but worthwhile in the long run

Next steps

Takeaways and next steps

• https://indico.cern.ch/event/1527699/contributions/6567328/attachments/3130748/5553910/OSCARS_Composability_FutureTasks.pdf