These scripts are used to help with monitoring several SpEC BhNs runs on a cluster which use Hydro-AMR. At the top of each script, there's a usage description. At this time, only per-segment-volume-generator.sh requires a command line argument.

Before you begin

1. git clone git://github.com/brege/bhns-scripts.git

2. cp sample.conf paths.conf

3. Edit paths.conf to provide (absolute) simulation paths to an epoch directory. E.g. /path/to/my/simulation/Lev1_Plunge or $SCRATCH/my/simulation/Lev1_Plunge/Lev1_SettleDisk

Monitoring .dat files

• domain-info.sh: Uses SpEC tool DomainInfo to get number of points and subdomains over time

• gr-domain-info.sh: Same as above, just on the GR side

• keep-last-checkpoint.sh: Ran in four hour crontab, this cleans checkpoint directories but keeps the last one for each segment

• latest-times.sh: Prints a quick overview table of the simulations

• rejoin-segments.sh: Iterates over all simulations to do CombineSegments.py, so all *.dat's are joined together

Automating some parts of the process to visualize matter on the grid

• per-segment-volume-generator.sh: For a specified segment from a specified run, run ConvertToVtk to output data in paraview format

• volume-generater.sh: Loops through all simulations and dumps latest data for paraviewing. You must specify the level to output, e.g. volume-generater.sh Lev1. If you add many levels, this command may take a long time

• match-domain-time-to-pv-file.py: simple python script to do the arithmetic that bash cannot do. (Got all the way through the bash script before realizing this. Would recommend going back and doing this in python instead.)

TODO

• Figure out if there's a way to transition bash loops to a parallel method via MPI