-
Notifications
You must be signed in to change notification settings - Fork 11
MiniRun5 file locations
https://portal.nersc.gov/project/dune/data/2x2/simulation/productions/MiniRun5_1E19_RHC/
MiniRun5 is a 1E19 POT sample used mainly as a testbed for the latest improvements to larnd-sim and flow, and as a precursor to "the big one". The edep-sim and Minerva outputs are frozen, but there will likely be multiple iterations of the larnd-sim and flow outputs, mlreco and CAFs.
-
"
beta1
": Failure rate: 2 runs out of 1024 for the whole chain -
We are currently at
beta2a
version of the run, current failure is 142 runs our of 1024 for the whole chain. Failure are at Flow level and seem to be related to flash finder and is under investigation.
Simulation update notes for MiniRun5 compared to MiniRun4:
- File naming convention:
- .ghep.root ->.GHEP.root
- gtrac->GTRAC
- .h5->.hdf5
- Improved detector geometry (https://github.com/DUNE/2x2_sim/tree/develop/geometry/Merged2x2MINERvA_v4). Now the 2x2 center is at (0, 0, 0) in (x, y, z) instead of (0, -268, 1300) cm in MiniRun4. The gap between the last upstream Mx2 and first downstream Mx2 planes is 3.5 m.
- Interaction ID renumbering so it won't skip events that don't have segments in 2x2 or Mx2. Add the rock interaction indicator in run number / run id to GENIE (GHEP) which further forms interaction ID / vertex_id . Previously, this information was only available from spill builder and onwards. See PicoRun4 and 4.1 notes.
Specific beta2b
updates are listed in those presentations:
-
For Larndsim and Flow https://docs.dunescience.org/cgi-bin/sso/RetrieveFile?docid=30560&filename=2x2_analysis_meeting_16May2024.pdf&version=2
-
For MLreco and CAFs https://docs.dunescience.org/cgi-bin/sso/RetrieveFile?docid=30569&filename=CAFissues_NDprotoypes_May16.pdf&version=1
/exp/dune/data/users/noeroy/prod/MiniRun5_1E19_RHC/MiniRun5_1E19_RHC.caf.beta1/
/exp/dune/data/users/noeroy/prod/MiniRun5_1E19_RHC/MiniRun5_1E19_RHC.caf.beta2a/
/global/cfs/cdirs/dune/www/data/2x2/simulation/productions/MiniRun5_1E19_RHC/MiniRun5_1E19_RHC.caf.beta1
/global/cfs/cdirs/dune/www/data/2x2/simulation/productions/MiniRun5_1E19_RHC/MiniRun5_1E19_RHC.caf.beta2a
Currently (as of beta1
) we're publishing the flow, mlreco, MINERvA, and GENIE files (i.e. the direct inputs to the CAF maker):
-
/exp/dune/data/users/noeroy/prod/MiniRun5_1E19_RHC/
for Minerva and MLReco -
/pnfs/dune/persistent/users/noeroy/prod/MiniRun5_1E19_RHC/
for the Flow and Larndsim
/global/cfs/cdirs/dune/www/data/2x2/simulation/productions/MiniRun5_1E19_RHC/
Available at NERSC, but be advised that these are liable to get overwritten during later iterations:
/pscratch/sd/d/dunepr/output/MiniRun5
- https://portal.nersc.gov/project/dune/data/2x2/simulation/productions/MiniRun5_1E19_RHC/failure_logs_beta1_final/
- https://portal.nersc.gov/project/dune/data/2x2/simulation/productions/MiniRun5_1E19_RHC/failure_logs_beta2a/
- beta2a
https://github.com/DUNE/larnd-sim/tree/MiniRun5-beta2
https://github.com/DUNE/ndlar_flow/tree/MiniRun5-beta2
https://github.com/DeepLearnPhysics/edep2supera/tree/v1.3.1
https://github.com/DeepLearnPhysics/flow2supera/tree/v3.0.1
https://github.com/DeepLearnPhysics/lartpc_mlreco3d/tree/v2.9.5
https://github.com/DUNE/ND_CAFMaker/tree/v4.6.3
MiniRun production that varied the ionization potential (W_ion) parameter in larnd-sim to +/-3 sigma (to 25.1 and 22.7 eV respectively) from its nominal value (23.6 eV). The GENIE and edep-sim inputs are the same as MR5 beta1 and produced new files for larnd-sim and the rest of the simulation chain, including CAFs. This production was primarily run at ANL, but for some technical difficulties the CAF stage was performed at NERSC.
Files are available at NERSC and FNAL. In each directory below there is a sub-directory for each stage of the 2x2 simulation (if present on the cluster). For example, CAFs can be found in MiniRun5_Systematics_W_ION_22.7.caf
.
/global/cfs/projectdirs/dune/www/data/2x2/simulation/productions/MiniRun5_Systematics_W_ION_22.7
/global/cfs/projectdirs/dune/www/data/2x2/simulation/productions/MiniRun5_Systematics_W_ION_25.1
The directories on FNAL do not contain the GENIE or edep-sim inputs; see the MR5 beta1 directories for those inputs.
/pnfs/dune/persistent/users/cuddandr/2x2_production/MiniRun5_Systematics_W_ION_22.7
/pnfs/dune/persistent/users/cuddandr/2x2_production/MiniRun5_Systematics_W_ION_25.1
Inside the directories above they link to the original files located here except for CAFs
/pnfs/dune/persistent/users/fmaha/productions/MiniRun5_Systematics_W_ION_22.7
/pnfs/dune/persistent/users/fmaha/productions/MiniRun5_Systematics_W_ION_25.1