Changes between Version 7 and Version 8 of Ticket #1943
- Timestamp:
- 2018-10-11T11:56:39+02:00 (6 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Ticket #1943
- Property Cc timgraham removed
- Property Wp_comment changed from The improvement of the hybrid coarse-grained approach (about 7% when the MPI scalability limit is reached) is not enough to justify the refactoring of the code. A final test by combining a tessellated approach with OpenMP will be performed and the ticket will be closed by the end of this year. to The improvement of the hybrid fine-grained and coarse-grained approaches is under discussion within the NEMO HPC-WG. The action will be completed and the ticket will be closed by the end of 2018. No development branch to be included in NEMO 4.0
-
Ticket #1943 – Description
v7 v8 1 1 == Context == 2 Evaluation of the scalability improvement due to the introduction od a second level of parallelism in NEMO based on the MPI+X model (where X is OpenMP). A new coarse-grained approach is evaluated in order to improve the efficiency of the fine-grained version developed during 2016. 3 2 Evaluation of the scalability improvement due to the introduction of a second level of parallelism in NEMO based on the MPI+X model (where X is OpenMP). A new coarse-grained approach is evaluated in order to improve the efficiency of the fine-grained version developed during 2016. 4 3 5 4 == Implementation == 6 5 7 1. the new ZDF package, rewritten by Gurvan is considered as test case (2017) 8 2. a single OpenMP parallel region that includes all the ZDF packages (2017) 9 3. implementation of a module for the OpenMP horizontal decomposition; introduction of OpenMP directives within the ZDF manager (2017) 10 4. tests of restartability and reproducibility on GYRE_PISCES (2017) 11 5. definition of the computational performance test plan and performance evaluation on different HPC systems (2017) 12 6. discussion with the NEMO ST and HPC-WG (2017) 13 7. extension of the approach to the tracers advection (2018) 14 8. definition of the computational performance test plan and performance evaluation on CMCC HPC system (2018) 15 9. discussion with the HPC-WG (2018) 6 1. the new ZDF package, rewritten by Gurvan is considered as test case (2017) --> done 7 2. a single OpenMP parallel region that includes all the ZDF packages (2017) --> done 8 3. implementation of a module for the OpenMP horizontal decomposition; introduction of OpenMP directives within the ZDF manager (2017) --> done 9 4. tests of restartability and reproducibility on GYRE_PISCES (2017) --> done 10 5. definition of the computational performance test plan and performance evaluation on different HPC systems (2017) --> done 11 6. discussion with the NEMO ST and HPC-WG (2017) --> done 12 7. extension of the approach to the tracers advection (2018) --> done 13 8. definition of the computational performance test plan and performance evaluation on CMCC HPC system (2018) --> done 14 9. discussion with the HPC-WG (2018) --> ongoing