Changes between Initial Version and Version 1 of Ticket #851


Ignore:
Timestamp:
04/27/20 22:27:19 (4 years ago)
Author:
arango
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #851

    • Property Resolution → Done
    • Property Status new → closed

  • Ticket #851 – Description

    initial v1  
    1 I have several research repositories for several ongoing new algorithms that will be available in the future. I have so many exceptional flags to update the codes that sometimes one fails and undesired changes appear in the public released version of ROMS and erase features in the research versions.  To avoid doing that in the future, I am consolidating the research 4DVAR repository with the public version of the code.
     1I have several research repositories for several ongoing new algorithms that will be available in the future. I have so many exceptional flags to update the codes that sometimes one fails, and undesired changes appear in public released version of ROMS and erase features in the research versions.  To avoid doing that in the future, I am consolidating the research 4DVAR repository with the public version of the code.
    22
    3 For the last two months, we have been developing the saddle-point 4D-Var. By far, one of the most technically tricky algorithms that Andy and I have put our hands on.  As formulated, the current available 4D-Var algorithms are sequential, and its inner loops cannot be parallelized in time to accelerate the analysis. The conjugate gradient solver, used to minimize the cost function '''J''', depends sequentially on the previous iteration.
     3For the last two months, we have been developing the saddle-point 4D-Var. By far, one of the most technically tricky algorithms that Andy and I have put our hands on.  As formulated, the current available 4D-Var algorithms are sequential, and its inner loops cannot be parallelized in time to accelerate the analysis. The conjugate gradient solver used to minimize the cost function '''J''', depends sequentially on the previous iteration.
    44
    55Fisher and Gürol (2017) proposed a new weak constraint 4D-Var algorithm based on the saddle-point formulation that can be parallelized in time. The time trajectory can be split into the solution of '''Nsaddle''' intervals that can run concurrently to reduce the 4D-Var computational cost.
    … …  
    1313* Include several new generic routines to manipulate ROMS state vector: '''ADfromTL.F''', '''def_state.F''', '''state_join.F''', '''state_read.F''', and '''wrt_state.F'''.
    1414
    15 * No change is made to ROMS nonlinear, tangent linear, and adjoint time-steping kernels.
     15* No change is made to ROMS nonlinear, tangent linear, and adjoint time-stepping kernels.
     16
     17* Added several routines from the LAPACK library that are needed for the Generalized Minimal Residual ('''GMRES''') minimization solver.
    1618
    1719* The saddle-point algorithm with be released in the future when it is fully tested and we have the proper documentation and relevant publication.