Gulf Stream Finder Execution


The source codes, scripts, executables, and static data files are in ~rmg3/gsfinder/ on ursus-marinus.

Execution

ctl.sh -- is the main control script.  It invoked executables parse.pl, gsfinder, and form (in that order), as well as managing data collection and distribution.  

parse.pl -- Reads the navy analysis (if there is one) and translates it to a format that can be read more easily by the Gulf Stream finder.

gsfinder -- This is the major program in the process.  
  1. It reads in the analyzed Gulf Stream if there is one, then evolves the best function for finding the Gulf Stream in ROFS output and prints out that result.  If there is no analysis available, it will use the previous best algorithm to produce the Gulf Stream file.
  2. It reads in the previous best algorithm if there is one and continues evolution from there.
  3. If there is no previous best algorithm and there is an analysis file, it will evolve a new population.
  4. If there is no previous algorithm and there is no analyzed Gulf Stream file, this will be a probably fatal problem for the program.
form -- Takes the output of gsfinder and formats it for reading by NAWIPS stations for use by OPC.


Local Data Files

Static files -- (in ~rmg3/gsfinder/)
    cfs.bini, cfs.binj, cfs.bin.full are static files describing the native ROFS grid.

Files saved from day to day -- (in ~rmg3/disk1/gsfout)
    genes_out.YYYYMMDD  -- the genome set for the 8 digit date YYYYMMDD
    genes_in -- the input set to use the next time there is an analysis available.  It will match the genes_out file for the most recent analysis day.
    -- the files are split so as to avoid needing to do date arithmetic in ctl.sh to find the most recent day.
    -- The split also permits manual editing of the genes_in file without erasing the record of solutions found.

Source Files

(in ~rmg3/gsfinder/source)



Return to main Gulf Stream Finder page
Return to main MMAB page

Last Modified 13 March 2003
Robert Grumbine

Valid HTML 4.01!