Setting Up A New AstroBEAR Directory

First things first, move to a directory of your choice and type

svn co svn+ssh://[username]@clover.pas.rochester.edu/data/repositories/doc Documentation

This will create a new directory Documentation/ in the current directory, and it will check out from Subversion (SVN) a number of .tex files that will be used to set up the new problem directory. This is done automatically when you check out the contents of the doc folder.

Next, navigate into the Documentation directory. In it, you will find the firstrun/ directory, which can be copied, contents and all (cp -r firstrun DEST_PATH) to any location you like. This is the sample problem directory, and can be used as a template for others.

You will notice that firstrun/ contains two subdirectories: out/ and TABLES/. out/ is the directory to which AstroBEAR writes its data files (notably the Chombo HDF5 files used by our visualization packages. TABLES/ contains various tables used by astroBEAR for different problems, i.e cooling and shocks. In addition, firstrun/ will contain template.shockclump.f90. The executable you use should be compiled using template.f90 pointing to this file.

Your firstrun/ directory contains the following .data files:

• domain.data: The main data file, this contains the information needed to manage the global domain parameters.

• physics.data: Contains constants and switches used by various physics modules, such as scaling factors and source term flags. The lMHD flag that indicates an MHD problem is also located here and so is the cooling flag.

• problem.data: Contains the switches for choosing the module for astrobear to run. When template is switched to T, such as it is in firstun/, astrobear looks for template.data for the problem setup. However, if the problem is another module (i.e. lClumps), astrobear will look for that module's .f90 file for specification of that particular problem. Make sure the appropriate module files are located in the directory from which astrobear is being run for your particular problem setup.

• template.data: Contains some constants for clump and shock problems. Note that these constants are in computational units rather than physical units. This file may change depending on the particular module you are running.

All of these files are set up as part of the SVN checkout process, so users need only modify them to suit later problems.

The final thing to do is to copy the executable (cp ASTROBEAR_DIR/xbear DEST_PATH/xbear for single-processor jobs, cp ASTROBEAR_DIR/mpibear DEST_PATH/mpibear for multi-processor jobs) to the new problem directory. Once you do this, you should be ready to run your simulation.

At the end of the simulation you may encounter an error complaining about the absence of a "run.report". This does not affect the simulation, however, if you'd like to see this error go away you can simply make a file in the problem directory called run.report.