Geant4 Geometry and Transport Working Group Meeting

Europe/Zurich
1/1-025 (CERN)

1/1-025

CERN

20
Show room on map
Zoom Meeting ID
66653659333
Host
John Apostolakis
Useful links
Join via phone
Zoom URL
    • 14:30 15:20
      Part 1 - VecGeom: Plans for 2026 - open issues, identifying priorities and 'top-level' deliverables

      POW VecGeom 2026: restructure VecGeom to deliver higher GPU navigation performance, consistent behavior across backends, and a maintainable, portable code base (CUDA/HIP). The detailed plan is drafted here

      • Simplify and standardize the navigation interface (clear inputs/outputs; consistent boundary behavior; consistency of VNavigator/GPU interfaces).

      • Improve GPU navigation performance with a more locality-aware accelerator (beyond today’s BVH)

      • Replace recursion-heavy algorithms (esp. Boolean navigation) with iterative, stack-friendly implementations

      • Remove legacy vector/SIMD remnants in kernels to simplify code and improve scalar performance

      • Add a portability layer for CUDA/HIP and refactor device code toward POD + explicit dispatch (no device virtuals)


      Mid-term items:

      • Mixed-precision implementation (global API in double; per-solid precision policies)
      • Documentation: user guides + algorithm notes to match the restructured core

       

      Long-term goal: Unified solids common library providing solid-level and global navigation algorithms for Geant4, ROOT, reconstruction programs.

       

    • 15:20 15:30
      Short break 10m
    • 15:30 16:50
      Part 2. Other Topics: Status and Plans for 2026
      • 15:30
        Status and Outlook 10m
        Speaker: John Apostolakis (CERN)
      • 15:40
        Goals for 2026 50m

        Recall topics mentioned in 2025:

        Navigation / modelling
        - Vecgeom surface-based modeller
        Performance / interfaces
        - Improvement of implementation of parallel voxel creation
        - Overlap checking using Multi-Threading
        Field:
        - Robustness improvement of QSS3
        - High order symplectic stepper (needs person)

        Impact / Coupling with other developments (in Geometry & Transport and beyond)

        See Deliverables (2024 & 2025) and potential deliverables (future)

      • 16:30
        Reports, Open Topics and Plans - roundtable 20m
    • 16:50 17:00
      AOB: Other topics