Skip to content

Coverage and known gaps

This site documents six sim/reco workflow presentations to the Phlex Adoption Working Group. The pages overlap in detector scope and do not cover all DUNE software.

Last verified: 2026-08-26.

Current coverage

Area Coverage Evidence
protoDUNE data hierarchy Partial DAQ terminology and offline object chain; no end-to-end production deck
ND prototypes Partial Four-slide hierarchy and workflow overview
ND-LAr and TMS sim/reco Strong Full production-chain deck
SAND sim/reco Strong Full workflow deck, framework status, and capacity concern
FD TPC sim/reco Strong Module/product chain, hierarchy proposal, and migration entry points
FD PDS sim/reco Strong Module/product chain, bottleneck, and migration recommendation
Existing FD art/FHiCL entry points Partial Literal configurations are documented for the FD TPC and PDS talks only
DUNE-native Phlex pathfinder packages Partial dune-xerosere package graph and reported status
Migration inventory Dated baseline April 2026 static audits, not current execution status

Material gaps

Core software and detector services, analysis and validation, and trigger/DAQ/raw-data code account for 258 of the 433 direct declarations in the April 2026 scope estimate. These areas have limited coverage here; the estimate also lacks a public, reproducible scan reference.

Pages or source-backed coverage are still needed for:

  • conditions, calibration, geometry, channel maps, and framework services;
  • DAQ ingress, trigger decisions, and art::TriggerResults replacement requirements;
  • FORM, RNTuple, HDF5, schema evolution, and persistent references;
  • analysis and CAF consumers;
  • production orchestration, justIN, validation, and regression comparisons;
  • core/common plugin migration and ownership;
  • actual milestone progress, staffing, and current blockers.

The I/O, persistence, and associations page records design requirements. Supported configurations and executable validation results are still needed.

Evidence and maintenance gaps

The following gaps remain after the September 2026 documentation review. This review did not reverify implementation status.

Gap Evidence needed
Source access Public links and slide/page references for the six workflow presentations, the July pathfinder report, the FORM talk, and the CDR conclusions
Migration inventory Public scan inputs, repository revisions, counting rules, and results supporting the April totals and figures
Current progress An owner-reviewed F2 update, repository status, blockers, and a dated public tracker
Executable validation A pinned software environment, sample inputs, commands, expected outputs, and comparison tolerances for the first migrated workflow
FHiCL discrepancy A release-specific check of the conflicting Reco2 filenames in the FD TPC and PDS presentations
Convention decisions Public proposals, review dates, and recorded adoption decisions for units, timestamps, and Python
Ownership Confirmed maintainers and review contacts; presenting a workflow does not establish responsibility for its migration

How to interpret the six workflow pages

Treat each page as a documented workflow scope:

  • explicit: stated directly in that scope's source deck;
  • inherited: supplied by a companion presentation covering the same production chain;
  • inferred: a cross-source interpretation that needs owner validation;
  • unaddressed: the source does not provide evidence either way.

This evidence vocabulary is used in the hierarchy comparison to avoid turning missing discussion into apparent consensus.