A Scalable Infrastructure for LHC Open Data Analysis and FAIR Data Provisioning

by Dr Baida Achkar (Georg-August-Universitaet Goettingen - II. Physicalishes Institut)

Thursday 18 Sept 2025, 12:30 → 13:30 Europe/Berlin

 

 

abstract:

The EXPLORE service, developed within the PUNCH4NFDI consortium in TA2 and offered as service by TA6,  provides a scalable and accessible infrastructure for analysing Large Hadron Collider (LHC) Open Data in line with FAIR (Findable, Accessible, Interoperable, Reusable) principles. Hosted at the GoeGrid cluster in Göttingen, EXPLORE lowers barriers to high-energy physics (HEP) data analysis by eliminating the need for CERN or institutional credentials, specialized local setups, or dedicated hardware.

Through ready-to-use containerized environments, dynamic resource management, and batch processing with HTCondor, EXPLORE enables users worldwide, including high-school students, educators, and unaffiliated researchers, to run real ATLAS Open Data analyses at scale. Remote data access is seamlessly preconfigured, ensuring reproducibility and efficiency.

A showcase analysis demonstrates the platform’s capability: a top–antitop (tt̄) production analysis using the ATLAS Open Data 2025 Beta release (36 fb⁻¹ at √s = 13 TeV). The analysis successfully processed approx. 97 million events across 58 parallel jobs, validating EXPLORE’s performance, scalability, and suitability for large-scale workflows.

Since its deployment in 2023, EXPLORE has evolved from prototype to public adoption, now supporting education, outreach, and independent research. It complements existing interactive ATLAS Open Data tools (e.g. Jupyter notebooks) by providing compute-heavy batch analysis capabilities. Early adoption in Göttingen HEP masterclasses and dissemination efforts (ATLAS Week, CoRDI 2025, outreach to schools) underline its growing role as a bridge from Open Data in theory to Open Science in practice.

Key Impact: EXPLORE empowers global participation in HEP by making real analyses accessible beyond institutional boundaries, serving as both a training platform and a research-grade environment.

For access and tutorials: https://punchlogin.goegrid.gwdg.de

==============================================

Connection details:
ZOOM Meeting “PUNCHLunch seminar”:

https://desy.zoom.us/j/91916654877
Webinar ID: 919 1665 4877, passcode: 481572

20250918.PUNCHLunch-EXPLORE.pdf

EXPLORE-PUNCHLunches.pdf