Difference between revisions of "Main Page/PIO"
| Line 7: | Line 7: | ||
== Usage Introduction: == | == Usage Introduction: == | ||
| − | + | For traditional approaches based on one file per processor, users can use #1,2,3 | |
| − | + | in .rea file for a specific example, by parameter 81. <br> | |
| + | These three approaches produce one file per processor and we will refer them as ''old libraries''. <br> | ||
| + | |||
| + | '''param(81) = 1''': use nek5000 old_output format *.fld (ASCII). <br> | ||
| + | '''param(81) = 2''': use Fortran I/O library (ASCII, VTK format). <br> | ||
| + | '''param(81) = 3''': use C-POSIX I/O libraries (binary, VTK format). <br> | ||
| − | |||
| − | |||
| − | |||
| − | |||
---- | ---- | ||
| − | '''param(81) = 4''' | + | Several advanced parallel I/O approaches were developed. The details are described[]. |
| + | #4, 5, 6, (-6), 8 indicate which I/O to use, as follows: <br> | ||
| + | |||
| + | |||
| + | '''param(81) = 4''': collective for N processors IO to 1 file (mpi-binary-N1.vtk) <br> | ||
'''param(81) = 5''' is NM binary case: collective IO for N processors to multiple M-files. | '''param(81) = 5''' is NM binary case: collective IO for N processors to multiple M-files. | ||
Revision as of 10:30, 12 June 2011
This is the document page for parallel I/O library developed for NekCEM.
Background
- File Format
- Binary (used for production, compact size), or ASCII (used for debugging, human-readable)
Usage Introduction:
For traditional approaches based on one file per processor, users can use #1,2,3
in .rea file for a specific example, by parameter 81.
These three approaches produce one file per processor and we will refer them as old libraries.
param(81) = 1: use nek5000 old_output format *.fld (ASCII).
param(81) = 2: use Fortran I/O library (ASCII, VTK format).
param(81) = 3: use C-POSIX I/O libraries (binary, VTK format).
Several advanced parallel I/O approaches were developed. The details are described[].
- 4, 5, 6, (-6), 8 indicate which I/O to use, as follows:
param(81) = 4: collective for N processors IO to 1 file (mpi-binary-N1.vtk)
param(81) = 5 is NM binary case: collective IO for N processors to multiple M-files.
param(103) = 6 is NM1 binary case: reduced-blocking IO for N processors to 1 file with M writers.
param(103) = 8 is NMM binary case (rbIO to M files with M writers).
param(103) = -6 is NM1 ASCII case (in tests it's refered to "7" instead of "-6").
Note that param(82) and param(83) need to be set correctly in *.rea file.
param(82) = number of groups, param(83) = max number of fields we are going to write.
