Changes between Version 30 and Version 31 of Doc/env/TgccCurie


Ignore:
Timestamp:
05/17/16 17:10:56 (8 years ago)
Author:
mafoipsl
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • Doc/env/TgccCurie

    v30 v31  
    4747PARTITION    STATUS   TOTAL   DOWN    USED    FREE    TOTAL   DOWN    USED    FREE     MpC  CpN SpN CpS TpC 
    4848---------    ------   ------  ------  ------  ------  ------  ------  ------  ------   ---- --- --- --- --- 
    49 standard     up        80368      32   77083    3253    5023       1    4824     198   4000  16   2   8   1 
    50 xlarge       up        10112     128    1546    8438      79       1      14      64   4000  128  16   8   1 
    51 hybrid       up         1144       0     264     880     143       0      33     110   2900   8   2   4   1 
     49standard     up        80352       0   76161    4191    5022       0    4724     298   4000  16   2   8   1 
     50hybrid       up          264       0       0     264      33       0       0      33   2900   8   2   4   1 
    5251}}} 
    5352 * detail of a running job. One command per line ccc_mprun : 
    … …  
    137136}}} 
    138137 
    139 # Fat nodes / Thin nodes # 
    140  
    141 As gencmip6 project has resources on standard node only, ins_job (libIGCM/ins_job since libIGCM_v2/7) ask you questions to know you project and where your post-processing jobs will run. Read carefully questions and default answers (frequently appropriate). 
    142  
    143 Thin nodes are two times faster than fat nodes for computations; they are as fast as fat nodes for post processing. 
    144  
    145 We decided to use thin nodes for computations and fat nodes for post processing.  
    146  
     138# Thin nodes # 
     139 
     140Since April 2016, only thin nodes are available at TGCCC.  
    147141The job header must include  #MSUB -q standard to use thin nodes. 
    148  
    149 The job header must include  #MSUB -q xlarge to use fat nodes. 
    150142 
    151143## SSD on standard node : how to use it for rebuild job ## 
    … …  
    156148#MSUB -x  # exclusive node 
    157149RUN_DIR_PATH=/tmp/REBUILD_DIR_MR_$$ 
    158 }}} 
    159  
    160 ## RAM on fat node : how to use it for rebuild job ## 
    161 If /tmp of standard node (64GB) is too small for you, you can use fat node. You have to know the /tmp of fat node is also of 64GB. You have to use the RAM disk which is 250GB/node. You have only to change header and RUN_DIR_PATH in rebuild.job. Take care you will run faster but cost will be multiplied by a factor of 128 because fat node ie 128 cpus are dedicated. 
    162  
    163 {{{ 
    164 #MSUB -q xlarge # fat nodes 
    165 #MSUB -x  # exclusive node 
    166 RUN_DIR_PATH=/dev/shm/REBUILD_DIR_MR_$$ 
    167150}}} 
    168151