Difference between revisions of "Main Page/PIO"
Line 8: | Line 8: | ||
Generally, several new parallel I/O approaches were developed, namely #3, 4, 5, 6, 7, 8. | Generally, several new parallel I/O approaches were developed, namely #3, 4, 5, 6, 7, 8. | ||
− | These numbers were used to indicate which I/O to use, in .rea file for a specific example, by parameter | + | These numbers were used to indicate which I/O to use, in .rea file for a specific example, by parameter 81. |
− | '''param( | + | '''param(81) = 1''' means it's using nek5000 output format *.fld (ASCII). <br> |
− | + | '''param(81) = 2''' mean it's using Fortran I/O library to write ASCII files in VTK format. <br> | |
− | '''param( | + | '''param(81) = 3''' means it's using C-POSIX I/O libraries to write binary files. <br> |
+ | These three approaches produce one file per processor and we will refer them as ''old libraries''. <br> | ||
---- | ---- | ||
− | '''param( | + | '''param(81) = 4''' is N1 binary case: collective IO to 1 file. |
− | '''param( | + | '''param(81) = 5''' is NM binary case: collective IO for N processors to multiple M-files. |
− | '''param(103) = 6''' is NM1 binary case | + | '''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) = 8''' is NMM binary case (rbIO to M files with M writers). |
Revision as of 10:20, 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:
Generally, several new parallel I/O approaches were developed, namely #3, 4, 5, 6, 7, 8. These numbers were used to indicate which I/O to use, in .rea file for a specific example, by parameter 81.
param(81) = 1 means it's using nek5000 output format *.fld (ASCII).
param(81) = 2 mean it's using Fortran I/O library to write ASCII files in VTK format.
param(81) = 3 means it's using C-POSIX I/O libraries to write binary files.
These three approaches produce one file per processor and we will refer them as old libraries.
param(81) = 4 is N1 binary case: collective IO to 1 file.
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.