Version 9 (modified by lovato, 8 years ago) (diff) |
---|
CMCC-NEMO - 2016 actions
Last edited Timestamp?
WARNING page currently in progress
members
Name | Institution | % | Position |
Dorotea Iovino | CMCC | 40 | NEMO Officer |
Stefania Ciliberti | CMCC | 20 | Scientist |
Tomas Lovato | CMCC | 20 | Scientist |
Silvia Mocavero | CMCC | 20 | Scientist |
Andrea Storto | CMCC | 0 | Scientist |
Summary: List here the actions NOT listed in SHARED ACTIONS
Action | Brief description | Status | Branch name | Trac ticket n° | Wiki page name | Reviewer(s) | Review status |
CMCC-6 | Remove key_bdy macro | Gurvan Madec |
Summary of SHARED ACTIONS (listed here for continuity)
Action | Brief description | Status | Branch name | Trac ticket n° | Wiki page name | Reviewer(s) | Review status | Shared Action Stream |
CMCC-1 | Scalability with MPI-OPENMP | |||||||
CMCC-2 | Suppress Global Comms (mpp_sum etc) | |||||||
CMCC-3 | BDY for nested limited area models | |||||||
CMCC-4 | BDY tool developement and tests | |||||||
CMCC-5 | Simplification | Gurvan Madec |
CMCC-1 - Scalability with MPI-OPENMP
Motivation: Improve NEMO scalability.
Status : continuation of 2015 dev CMCC-1
Main tasks : 1. the activity started at the end of 2013 with the analysis of three OpenMP approaches (implemented using “mini-apps”) on several multicore architectures; the second step was the implementation of the hybrid version of the GYRE configuration using the best OpenMP approach. Results in terms of scalability were not so good. Thus, the porting and the analysis of the hybrid version (using “mini-apps”) on many core architectures (e.g. Xeon Phi) is needed, before extending the hybrid OpenMP/MPI approach to the entire code, in order to justify the effort of the development. This analysis is carrying out in 2015. The extension of the hybrid implementation on the entire code will depend on the results of this analysis. 2. analysis of scalability improvement using MPI3 new communications (e.g. collective neighbours communications), instead of point to point communications
Science Reviewer:
System Reviewer:
Deadline:
Priority: HIGH
Depends on: results of the same action in 2015
Principal Investigator : S. Mocavero (silvia.mocavero@…)
CMCC-2 - Suppress global communications
Motivation: Removing global communications could help improve the speed of NEMO
Status : Not started
Main tasks : CMCC will investigate which communications could be safely removed. Met Office will look at a more efficient implementation of functionality in stp_ctl.
Science Reviewer:
System Reviewer:
Deadline:
Priority: HIGH
Depends on:
Principal Investigator : S. Mocavero (silvia.mocavero@…) and Tim Graham
CMCC-3 - BDY for nested limited area models
Motivation: Improving nesting techniques for regional and sub-regional circulation models nested into coarser models [BR]] Status :
Main tasks : 1. Implementation of the generalised Flather’s condition for the barotropic component (Oddo & Pinardi, 2008); 2. Interpolation Constraint applied to total velocity field for preserving the transport conservation across open boundaries
Science Reviewer: Prof. N. Pinardi (University of Bologna
System Reviewer:
Deadline:
Priority:
Depends on:
Principal Investigator : Stefania Ciliberti (stefania.ciliberti@…)
CMCC-4 - BDY tool developpement and tests
Motivation:
Status :
Main tasks : NERC will finish last improvement of the BDY tool and add some documentation, and CMCC will test the BDY tool
Science Reviewer:
System Reviewer:
Deadline:
Priority:
Depends on:
Principal Investigator : Stefania Ciliberti (stefania.ciliberti@…) and James Harle
CMCC-5 - Simplification
Motivation: Set the use of MPI safe computation as a default
Status :
Main tasks : remove the use of the macro key_mpp_rep and use as a default condition the MPI safe communications
Science Reviewer: Gurvan Madec
System Reviewer: Gurvan Madec
Deadline:
Priority:
Depends on:
Principal Investigator : Dorotea Iovino (diovino@…)
CMCC-6 - Remove key_bdy macro
Motivation: reduce the use of macros in NEMO
Status :
Main tasks : Remove the macro activating the code for the open boundary treatment in favour of static code always present in compilation and activated only when bdy are required by the user configuration
Science Reviewer: Gurvan Madec
System Reviewer: Gurvan Madec
Deadline:
Priority:
Depends on:
Principal Investigator : Tomas Lovato (tomas.lovato@…)