New URL for NEMO forge!   http://forge.nemo-ocean.eu

Since March 2022 along with NEMO 4.2 release, the code development moved to a self-hosted GitLab.
This present forge is now archived and remained online for history.
2012WP/2012Action_institutions_CMCC (diff) – NEMO

Changes between Version 10 and Version 11 of 2012WP/2012Action_institutions_CMCC


Ignore:
Timestamp:
2012-11-07T10:38:47+01:00 (10 years ago)
Author:
vichi
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • 2012WP/2012Action_institutions_CMCC

    v10 v11  
    3434 
    3535=== '''CMCC.1 Scalability Optimization''' === 
    36 '''          Motivation:''' The optimization of NEMO scalability. [[BR]] '''          Status:''' to be started in January 2012 [[BR]] '''          Main tasks:'''[[BR]] * Introduction of a new software layer in the NEMO stack software exploiting numerical optimized and parallel libraries such as  pBLAS, ScaLAPACK and PETSc. [[BR]] * Introduction of a new parallel approach based on the functional parallelism for the tracers and the momentum equations. The parallelism will be based on  spreading: (i) the computation of the tracers and momentum equations and (ii) the computation of the advection, Coriolis, pressure gradient and lateral diffusion terms. [[BR]] * Implementation of a new hybrid parallel approach for exploiting the 3D domain decomposition (MPI on the horizontal domain, OpenMP on the vertical levels) [[BR]] '''          System Reviewer:''' [[BR]] '''          Deadline:'''  2012[[BR]] '''          Priority:'''Medium [[BR]] '''          Principal investigator:''' Italo Epicoco [[BR]] 
     36'''          Motivation:''' The optimization of NEMO scalability. [[BR]] '''          Status:''' to be started in January 2012 [[BR]] '''          Main tasks:'''[[BR]] * Introduction of a new software layer in the NEMO stack software exploiting numerical optimized and parallel libraries such as  pBLAS, ScaLAPACK and PETSc. [[BR]] * Introduction of a new parallel approach based on the functional parallelism for the tracers and the momentum equations. The parallelism will be based on  spreading: (i) the computation of the tracers and momentum equations and (ii) the computation of the advection, Coriolis, pressure gradient and lateral diffusion terms. [[BR]] * Implementation of a new hybrid parallel approach for exploiting the 3D domain decomposition (MPI on the horizontal domain, OpenMP on the vertical levels) [[BR]]* Optimization of communications in the open boundary conditions[[BR]] '''          System Reviewer:''' [[BR]] '''          Deadline:'''  2012[[BR]] '''          Priority:'''Medium [[BR]] '''          Principal investigator:''' Italo Epicoco [[BR]] 
    3737 
    3838=== '''CMCC.2 '''         Modification of masks ===