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.
WorkingGroups/HPC (diff) – NEMO

Changes between Version 43 and Version 44 of WorkingGroups/HPC


Ignore:
Timestamp:
2018-06-25T15:32:57+02:00 (4 years ago)
Author:
mocavero
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • WorkingGroups/HPC

    v43 v44  
    44Last edition: '''[[Wikinfo(changed_ts)]]''' by '''[[Wikinfo(changed_by)]]''' 
    55 
    6 Working group leader (and responsible for wiki pages) : Mike Bell 
     6Working group leaders (and responsible for wiki pages) : Mike Bell & Silvia Mocavero 
    77 
    88---- 
    99 
    1010== Members of the Working group: 
     11 * Miroslaw Andrejczuk (Met Office/ST) 
     12 * Mike Bell (Met Office) 
     13 * Miguel Castrillo (BSC) 
     14 * Matt Glover (Met Office) 
     15 * Claire Levy (CNRS/NEMO Project Manager) 
     16 * Eric Maisonnave (CERFACS) 
     17 * Sébastien Masson (CNRS/ST) 
     18 * Francesca Mele (CMCC/ST) 
     19 * Silvia Mocavero (CMCC/ST) 
     20 * Andrew Porter (STFC) 
     21 * Oriol Tinto (BSC) 
     22 
     23'''Other members''' 
     24 
     25 * Lucien Anton (CRAY) 
     26 * Clement Bricaud (MERCATOR/ST) 
     27 * Marcin Chrust (ECMWF) 
     28 * Louis Douriez (ATOS) 
     29 * Peter Dueben (ECMWF) 
     30 * Marie-Alice Foujols (CNRS)  
     31 * David Guibert (ATOS) 
     32 * Michael Lange (ECMWF) 
     33 * Julien Le Sommer (CNRS) 
     34 * Gurvan Madec (CNRS/NEMO Scientific Leader) 
     35 * Kristian Mogensen (ECMWF) 
     36 * Stan Posey (NVIDIA) 
     37 * Martin Price (Met Office/ST) 
     38 * Erwan Raffin (ATOS) 
     39 * Martin Schreiber (Univ of Exeter) 
     40 * Nils Wedi (ECMWF) 
     41 
     42'''Former members''' 
     43 
    1144 * Jeremy Appleyards (NVIDIA) 
    12  * Lucien Anton (CRAY) 
    13  * Mike Bell (Met Office) 
    14  * Tom Bradley (NVIDIA)  
    15  * Miguel Castrillo (BSC) 
     45 * Tom Bradley (NVIDIA) 
    1646 * Mondher Chekki (Mercator-Ocean)  
    17  * Marie-Alice Foujols (CNRS)  
     47 * Tim Graham (Met Office) 
     48 * Jason Holt (NOC) 
     49 * Dmitry Kuts (Intel) 
     50 * Cyril Mazauric (ATOS) 
    1851 * Yann Meurdesoif (CEA) 
    19  * Tim Graham (Met Office) 
    20  * Matt Glover (Met Office) 
    21  * Jason Holt (noC) 
    22  * Dmitry Kuts (Intel)  
    23  * Claire Levy (CNRS) 
    24  * Gurvan Madec (CNRS) 
    25  * Sébastien Masson (CNRS)  
    26  * Cyril Mazauric (ATOS) 
    27  * Silvia Mocavero (CMCC) 
    28  * Andrew Porter (STFC) 
    29  * Stan Posey (NVIDIA) 
    30  * Martin Schreiber (Univ of Exeter)  
    3152 * Kim Serradell (BSC) 
    32  * Oriol Tinto (BSC) 
    33  * Julien le Sommer (CNRS) 
     53 
     54 
     55== Main activities: 
     56 
     57'''NEMO benchmark''' 
     58 
     59''Description'': development of a benchmark (idealised geometry, equal MPI sub-domain size, realistic parameters, parametrisable communications, internal timings ...) at different resolutions to identify the bottlenecks to NEMO scalability. 
     60 
     61''Involved people'': Sebastien Masson (CNRS/ST), Eric Maissonave (CERFACS) 
     62 
     63''WP action'': HPC09_ESIWACE? 
     64 
     65'''Intra-node performance''' 
     66 
     67''Description'': analysis and improvement of NEMO intra-node performance to move real towards peak performance and to reduce intra-node communications. Analysis at kernel level to understand which kernels are limiting performance; improving utilisation of memory hierarchy; loop vectorisation. Investigation and integration of a second level of parallelism based on the shared memory paradigm. Use of OpenMP (fine and coarse-grained parallelisation). 
     68 
     69''Involved people'': Silvia Mocavero (CMCC/ST), Francesca Mele (CMCC/ST), Mike Bell (MetO), Miroslaw Andrejczuk (MetO/ST), Matt Glover (MetO) 
     70 
     71''WP actions'': WP2018-01_Silvia Mocavero_singlecoreperf, WP2018-02_Francesca_Mele_hybrid 
     72 
     73 
     74'''Inter-node communications''' 
     75 
     76''Description'': limiting the communication overhead through the reduction of the MPI communications number and frequency, the investigation of advanced MPI communications and techniques to overlap communication and computation. 
     77 
     78''Involved people'': Silvia Mocavero (CMCC/ST) 
     79 
     80''WP actions'': WP2018-03_Silvia Mocavero_globcomm, WP2018-04_Silvia Mocavero_mpi3, WP2018-05_AndrewC-extendedhaloes 
     81 
     82 
     83'''Separation of concerns''' 
     84 
     85''Description'': investigation of a light-DSL approach to apply DSL in NEMO without impacting on the NEMO coding structure rules. The approach consists in processing the NEMO code to create an internal representation compliant to the PSyclone tool and manipulating the intermediate code to perform PSyclone transformations.   
     86 
     87''Involved people'': Andrew Porter (STFC) 
     88 
     89''WP action'': - 
     90 
     91 
     92'''I/O''' 
     93 
     94''Description'': Investigation of techniques to reduce the I/O overhead which increases with the number of processors. Extension of the use of XIOS for reading/writing restart files.  
     95 
     96''Involved people'': Miroslaw Andrejczuk (Met Office/ST) 
     97 
     98''WP actions'': WP2018-06_andmirek-XIOSread, WP2018-07_andmirek_XIOSwrite 
     99 
     100 
     101'''Mixed-precision''' 
     102 
     103''Description'': optimization of the NEMO model by using a mixed precision approach. Study on the precision needed by the different processes in NEMO. Implementation of the mixed precision approach in NEMO. 
     104 
     105''Involved people'': Miguel Castrillo (BSC), Oriol Tinto (BSC) 
     106 
     107''WP action'': WP2018-08_Mixed_precision 
     108 
     109 
    34110 
    35111----