Home Resources for Co-Design Programs Parallel-file-io Parallel-file-io_parallel-ascii-multifile

Parallel File I/O (parallel-ascii-multifile)

Version's name: Parallel File I/O (parallel-ascii-multifile) ; a version of the Parallel File I/O program.
Repository: [home] and version downloads: [.zip] [.tar.gz] [.tar.bz2] [.tar]
Patterns and behaviours:

Sequential ASCII file I/O

Implemented best practices: Parallel multi-file I/O ·

This is the simplest parallel file I/O for ASCII files, where every process reads or writes to a unique file, i.e. the number of processes equals the number of files. This may be disadvantageous for some situations, e.g. for checkpointing if needing to restart a computation on a different number of processes.

The following pseudo-code shows the process for writing data.

get_my_process_id(proc_id)
filename = "file" + to_string(proc_id) + ".txt"
open_file(filename)
write_ascii_data_to_file(data, filename)
close_file(filename)

The process is reversed when reading data from file, as follows.

get_my_process_id(proc_id)
filename = "file" + to_string(proc_id) + ".txt"
open_file(filename)
read_ascii_data_from_file(data, filename)
close_file(filename)