REJOIN - Rejoin/Split Distributed Data

REJOIN was written to tidy up data at the end of a parallel computation. The problem was simple. Suppose that a logical array of NPX by NPY processors was used, and that the computation was associated with a physical array of NX_GLOBAL by NY_GLOBAL processors. The original physical array was divided up among the processors, and each handled its own piece. (There was some communication between pieces, via what amounted to internal boundary conditions). Once the computation was done, each processor wrote out its information to a file, whose name had the processor ID embedded in it. In order to conveniently graph or analyze the data, it was desired to rejoin all this data into a single file, as though one processor had handled it. This is what REJOIN does, for the data from a specific program.

Of course, no sooner was code written to rejoin data from a logical array of NPX by NPY processors, than a request was made for a new feature, namely the ability to reshape the data, that is, to write a new set of files that would correspond to a configuration of, say, MPX by MPY processors. This would allow you to run on one configuration on Tuesday, save the data, and resume the calculation on Wednesday on a different processor configuration. It turns out that the easiest (though not most efficient!) way to do this is simply to add a "split" option to the code, which splits up a serial data file into an arbitrary set of parallel data files. The reshape option is then carried out by "rejoining" the data into a serial file, and "splitting" it out, specifying a different processor configuration. See the SPLIT_SAVE routine for details.

Files you may copy include:

• rejoin.f90, the source code;
• rejoin_prb.f90, a calling program to REJOIN parallel data;
• rejoin_prb.out, the output from the REJOIN operation.
• split_prb.f90, a calling program to SPLIT serial data;
• split_prb.out, the output from the SPLIT operation.
• convert_prb.f90, converts unformatted data to formatted;
• field_l_0000_.txt, the first (of 8) formatted data file (DELETED, TOO LARGE);
• field_l.txt, the rejoined formatted data file (DELETED, TOO LARGE);

The list of routines includes:

• CONVERT_ALL converts all the files from unformatted to formatted form.
• CONVERT_ONE converts one file from unformatted to formatted form.
• DIGIT_TO_C returns the character representation of a decimal digit.
• I_TO_S_ZERO converts an integer to a string, with zero padding.
• PFILE_CLOSE closes all the "parallel" files.
• PFILE_NAME_GEN generates a file name that contains a processor ID.
• PFILE_OPEN_READ opens all the "parallel" files for reading.
• PFILE_OPEN_WRITE opens all the "parallel" files for writing.
• PFILE_READ_PDATA reads "parallel" data from the parallel files.
• PFILE_READ_SDATA reads "scalar data" from a set of parallel files.
• PFILE_WRITE_PDATA writes "parallel" data to the parallel files.
• PFILE_WRITE_SDATA writes "scalar data" to a set of parallel files.
• REJOIN_SAVE "rejoins" a set of parallel DNS SAVE files into one serial file.
• SFILE_CLOSE closes a serial file.
• SFILE_OPEN_READ opens a serial file for reading.
• SFILE_OPEN_WRITE opens a serial file for writing.
• SFILE_READ reads data from a serial file.
• SFILE_WRITE writes data to the serial file.
• SPLIT_SAVE "splits" a single serial DNS SAVE file into parcels.

Return to the FORTRAN software page.