Report to WLCG MB #336 on Tuesday, January 20, 2025 at 16:00 (CERN)
Pending Item from the Previous WLCG MB #335 (December 2025)
From the TCB report to the last WLCG MB #335 on December 16th, there was a comment from CMS on the summary report from the WLCG Workshop on Heterogeneous Architectures concerning the statement, “ATLAS and CMS foreseeing being able to offload 15% or less of their wallclock time of processing to GPUs by the start of Run 4.” and that CMS will provide a range for the GPU offload.
Reply from CMS (January 19th): “Recent estimates from the CMS Conceptual Design Report (CDR) process indicate that GPU acceleration could offload approximately 27–48% of the prompt reconstruction workload and 12–35% of the Monte Carlo simulation chain, depending on the success of ongoing R&D efforts. When combined across workflows, this corresponds to an overall GPU offload of roughly 15–40% of the total CMS computing workload, contingent on the achieved R&D outcomes.” [Ed: Emphasis added]
OTF#8
Indico: https://indico.cern.ch/event/1609088/
The agenda for OTF#8 (see link above), which will focus on tape systems performance and evolution, is well-developed. Thank you to the DOMA Conveners Katy Ellis and Johannes Elmsheuser for their hard work in preparing the agenda. OTF#8 will take place at CERN during the afternoon of Tuesday, February 3rd and all day on February 4th.
OTF#9
OTF#9 [https://indico.cern.ch/event/1598655/] will be co-hosted with the HEPIX Spring 2026 Workshop in Lisbon, April 20-24, and will focus on Facilities Evolution. We are in the process of defining the goals of the session with the HEPiX organizers before their call for abstracts opens later this week.
WLCG Technical Roadmap
We have created new e-groups for the facilitators of the various chapters e.g. wlcg-technical-roadmap-ch2-facility-evolution, or wlcg-technical-roadmap-facilitators-all. After the WLCG Workshop on Heterogeneous Architectures in December, we got two new volunteers to help facilitate Chapter 8 of the Roadmap on “New Architectures and Infrastructures”, Ianna Osborne and Doug Benjamin, who will join Oxana Smirnova.
We need to schedule a check-in meeting of the TCB in order to checkpoint progress on the various chapters. Our goal is to have a workable draft or the majority of the chapters to socialize by the time of the CHEP26 Conference in Bangkok in late May. The TCB has submitted an abstract to the Conference and we are hoping for a plenary presentation.
Text of the Abstract:
WLCG Technical Evolution: Preparing for HL-LHC with a Community-Driven Roadmap
Authors: The WLCG Technical Coordination Board: D. Benjamin, L. Betev, C. Bozzi, D. Britton, I. Chakaberia, B. Couturier, A. Dewhurst, A. Di Girolamo, K. Ellis, J. Elmsheuser, J. Flix, J. C. Luna, M. Litmaath, S. Lammel, J. Letts, M. Mascheroni, S. McKee, A. Melo, E. Moyse, I. Osborne, S. Piano, E. Sexton‑Kennedy, O. Smirnova, D. South, N. Srimanobhas, F. Stagni, and J. van Eldik
With the end of Run 3 approaching, the Worldwide LHC Computing Grid (WLCG) is entering an important transition toward the HL-LHC era. To meet the substantial increase in data volume, computational requirements, and resource heterogeneity, while preserving reliability, sustainability, and community cohesion, we have launched the development of the WLCG Technical Roadmap 2026-2030, a consensus-driven, community-owned plan that identifies key gaps and defines concrete milestones across the infrastructure and services landscape.
The roadmap outlines the major areas of evolution structured into nine chapters covering core areas such as data transfers, networking, facilities, authorization (tokens) & security, workflow management, and heterogeneous resources, including GPUs, High-Performance Computing, and Cloud. The coordination for these areas is handled by dedicated "Chapter Facilitators" within the WLCG Technical Coordination Board. Equally important is the evolution of the WLCG facilities themselves, covering compute provisioning, storage architectures (both disk and tape), networking capabilities, and the operational models required to support HL-LHC performance and reliability.
The roadmap also outlines the main challenges ahead - technical, organizational, and community-related and proposes mechanisms to address them through coordinated development, shared milestones, and strengthened collaboration across experiments and facilities. We will present the structure, methodology, and mid-term action plan for the next couple of years together with the expected path toward full implementation.
By presenting this roadmap at CHEP, we seek to engage the broader HEP community to gather feedback, foster collaboration, and build collective awareness. As one of the pillars of LHC physics and a cornerstone for data-intensive science, WLCG’s future relies on coordinated innovation and active community participation.
Other Matters
We need to get back to the XRootD Developers to see how the TCB can help with stakeholder outreach, perhaps in the context of the OTF meeting series.
