Campaign 2026-07

This page documents the 2026-07 meson structure simulation campaign.

(!) For the list of files go to DATA PAGE

Overview

The July 2026 campaign introduces the new 9xN energy scheme in compliance with the EIC early-science program (the proton beam moves from 10/18 GeV to a single 9 GeV electron beam against 100/130/275 GeV protons, plus the low-energy 5x41). It is a single-flavor production run: one afterburned signal → DD4hep → EICrecon, with CSV tables produced from both the DD4hep and reco outputs.

Energy ranges:

1. 5x41 GeV
2. 9x100 GeV
3. 9x130 GeV
4. 9x275 GeV

DIS parameters (from the event-generator inputs):

• x = 0.0001 - 1.0000
• q2 = 1 - 500

Statistics:

• 1000 events per file (note: smaller per-file statistics than the 5000-evt/file 2025/2026 campaigns), file index NNNN, ~1000 files per energy.

(!) Unlike campaign 2026-06, this campaign has no -official / -saveall / -background / -stv variant split — reconstruction is run directly on the single dd4hep output. For machine-background-mixed data see campaign 2026-06 and the Data page.

Software stack

Built and run with the official EIC image from cvmfs:

/cvmfs/singularity.opensciencegrid.org/eicweb/eic_xl:26.06-stable

Data layout (ifarm)

Base directory:

/work/eic3/users/romanov/meson-structure-2026-07

Subdirectories:

• eg-original-2026-06 — original event-generator ROOT files (one per energy, k_lambda_crossing_*.root, ~19 GB)
• eg-split — event-generator files split into 1000-event chunks (~20 GB)
• eg-hepmc — split files converted to HepMC (<energy>, <energy>-priority, <energy>-rest; ~2.3 GB)
• afterburner — event-generator output with crossing-angle + beam effects (<energy>-priority, *.hepmc3.tree.root + *.hist.root; ~6.5 GB)
• dd4hep — DD4hep full simulation (*.edm4hep.root; ~5.0 TB)
• reco — EICrecon reconstruction of dd4hep (*.edm4eic.root; ~2.2 TB)
• csv_dd4hep — acceptance / MC-truth CSV tables from dd4hep (~21 GB)
• csv_reco — reconstruction CSV tables from reco (~1.2 GB)
• analysis — analysis outputs (acceptance, eg-kinematics, gary_scripts, multicalo-lambda, …)

Per-energy .root file counts and sizes (from the ifarm disk):

Energy	dd4hep/ (.edm4hep.root)	reco/ (.edm4eic.root)
5x41	998 files   (647 GB)	998 files   (472 GB)
9x100	1000 files (1.4 TB)	999 files   (718 GB)
9x130	1000 files (1.7 TB)	1000 files (421 GB)
9x275	1000 files (1.4 TB)	1000 files (630 GB)

Pipeline config

The campaign is driven by full-sim-pipeline/config-campaign-26-07.yaml:

base_dir: "/work/eic3/users/romanov/meson-structure-2026-07"
container: "/cvmfs/singularity.opensciencegrid.org/eicweb/eic_xl:26.06-stable/"
energies: ["5x41", "9x100", "9x130", "9x275"]
event_count: 1000

afterburner_input:  "${base_dir}/eg-hepmc/${energy}-priority"
afterburner_output: "${base_dir}/afterburner/${energy}-priority"

# DD4hep (afterburner output is the input)
dd4hep_input:  "${base_dir}/afterburner/${energy}-priority"
dd4hep_output: "${base_dir}/dd4hep/${energy}"

# CSV tables from DD4hep
csv_dd4hep_input:  "${dd4hep_output}"
csv_dd4hep_output: "${base_dir}/csv_dd4hep/${energy}"

# Reconstruction runs on the dd4hep output
eicrecon_input:  "${dd4hep_output}"
eicrecon_output: "${base_dir}/reco/${energy}"
csv_eicrecon_input:  "${eicrecon_output}"
csv_eicrecon_output: "${base_dir}/csv_reco/${energy}"

Commands log

Event-generator split (ROOT → HepMC, 1000 events/file)

# EG originals (one file per energy)
/work/eic3/users/romanov/meson-structure-2026-07/eg-original-2026-06/k_lambda_crossing_0.000-5.0on41.0_x0.0001-1.0000_q1.0-500.0.root
/work/eic3/users/romanov/meson-structure-2026-07/eg-original-2026-06/k_lambda_crossing_0.000-9.0on100.0_x0.0001-1.0000_q1.0-500.0.root
/work/eic3/users/romanov/meson-structure-2026-07/eg-original-2026-06/k_lambda_crossing_0.000-9.0on130.0_x0.0001-1.0000_q1.0-500.0.root
/work/eic3/users/romanov/meson-structure-2026-07/eg-original-2026-06/k_lambda_crossing_0.000-9.0on275.0_x0.0001-1.0000_q1.0-500.0.root

# split each into 1000-event chunks (repeat per energy)
uv run 01_root_hepmc_klam_convert.py \
      --input-files .../eg-original-2026-06/k_lambda_crossing_0.000-9.0on275.0_x0.0001-1.0000_q1.0-500.0.root \
      --chunk-size 1000 \
      --events 10000000 \
      --output-prefix .../eg-split/9x275/msf_ev1000 \
      --events-per-file 1000

Full simulation pipeline

mkdir -p /work/eic3/users/romanov/meson-structure-2026-07

# cd where scripts are
cd /home/romanov/meson-structure-work/meson-structure

# sync python env
uv sync

# afterburner (crossing angle + beam effects)
uv run full-sim-pipeline/10_create_afterburner_jobs.py \
    --config full-sim-pipeline/config-campaign-26-07.yaml

# DD4hep full simulation
uv run full-sim-pipeline/20_create_npsim_jobs.py \
    --config full-sim-pipeline/config-campaign-26-07.yaml

# EICrecon reconstruction (runs on dd4hep output)
uv run full-sim-pipeline/30_create_eicrecon_jobs.py \
    --config full-sim-pipeline/config-campaign-26-07.yaml

# CSV converters (from dd4hep and from reco)
uv run full-sim-pipeline/40_create_csv_dd4hep_jobs.py \
    --config full-sim-pipeline/config-campaign-26-07.yaml
uv run full-sim-pipeline/42_create_csv_eicrecon_jobs.py \
    --config full-sim-pipeline/config-campaign-26-07.yaml

Each XX_create_*_jobs.py writes per-file container + SLURM scripts under <output>/jobs/ plus a top-level submit_all_slurm_jobs.sh / run_all_local.sh.