Changes between Initial Version and Version 1 of Ticket #6


Ignore:
Timestamp:
04/11/20 21:20:44 (5 years ago)
Author:
arango
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #6

    • Property Resolution → fixed
    • Property Status new → closed

  • Ticket #6 – Description

    initial v1  
    33 For example, if '''Nsaddle=4''', '''NtileI=2''', and '''NtileJ=2''', we will need '''16''' processes when '''CONCURRENT_KERNEL''' is off and '''32''' processes if it is activated.
    44 {{{
    5    mpirun -np 16 romsM roms.in      (CONCURRENT_KERNEL off)
     5   mpirun -np 16 romsM roms.in > & err &     (CONCURRENT_KERNEL off)
    66or
    7    mpirun -np 32 romsM roms.in      (CONCURRENT_KERNEL on)
     7   mpirun -np 32 romsM roms.in > & err &     (CONCURRENT_KERNEL on)
    88}}}
    99
    … …  
    1212* Fixed several logical errors in '''saddle_point.h''', '''atimev.F''', '''gmres.F''', and '''precsp.F''' to achieve identical convergence values for '''Jobs'''.
    1313
    14 Many thanks to Andy Moore for his immense help in tracking and fixing the algorithm code and logical bugs.
     14* Modified the following files:
     15 {{{
     16ROMS/Adjoint/ad_initial.F
     17ROMS/Drivers/is4dvar_ocean.h
     18ROMS/Drivers/saddle_point.h
     19ROMS/Drivers/w4dpsas_ocean.h
     20ROMS/Modules/mod_parallel.F
     21ROMS/Modules/mod_strings.F
     22ROMS/Utility/atimesv.F
     23ROMS/Utility/checkdefs.F
     24ROMS/Utility/convolve.F
     25ROMS/Utility/def_mod.F
     26ROMS/Utility/gmres.F
     27ROMS/Utility/inp_par.F
     28ROMS/Utility/obs_initial.F
     29ROMS/Utility/precsp.F
     30ROMS/Utility/read_phypar.F
     31ROMS/Utility/stdinp_mod.F
     32ROMS/Utility/strings.F
     33ROMS/Utility/tracing.F
     34ROMS/Version
     35}}}
     36
     37----
     38
     39Many thanks to Andy Moore for his immense help in tracking and fixing the algorithm code and logical bugs.