CRwiki

User Tools

Site Tools

Trace:
tcr_sri002

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
tcr_sri002 [2020/04/26 10:33] – tmatejuktcr_sri002 [2023/08/01 01:08] (current) – external edit 127.0.0.1
Line 35: Line 35:
  
 ==== sbatch example ==== ==== sbatch example ====
-Use Quantum Espresso 6.5. This description assumes that path to binaries is ''/home/users/${USER}/soft/qe-6.5/bin'' . Use bellow ''.sbatch'' file to run computation. +Use Quantum Espresso 6.5. This description assumes that path to binaries is ''/home/users/${USER}/soft/qe-6.5/bin'' . Use bellow ''qe-test.sbatch'' file to run computation. 
 +  #!/bin/bash -l
   #SBATCH --job-name="qe-test_N2_n32"   #SBATCH --job-name="qe-test_N2_n32"
   #SBATCH --nodes=2                   # number of computing_nodes   #SBATCH --nodes=2                   # number of computing_nodes
Line 47: Line 47:
      
   WORKDIR="/home/users/${USER}/soft_tests/qe_run_`date +%s`_${RANDOM}/"   WORKDIR="/home/users/${USER}/soft_tests/qe_run_`date +%s`_${RANDOM}/"
-  mkdir -p ${WORKDIR} +  mkdir -p ${WORKDIR} 
 +  cd ${WORKDIR} 
 +  
   export BIN_DIR="/home/users/${USER}/soft/qe-6.5/bin"   export BIN_DIR="/home/users/${USER}/soft/qe-6.5/bin"
   export PATH=${BIN_DIR}:$PATH   export PATH=${BIN_DIR}:$PATH
   export PSEUDO_DIR=${WORKDIR}   export PSEUDO_DIR=${WORKDIR}
   export TMP_DIR="/tmp"   export TMP_DIR="/tmp"
-   +     
-  mkdir -p ${WORKDIR} +
-  cd ${WORKDIR} +
   #copy input files and pseudo files to ${WORKDIR}    #copy input files and pseudo files to ${WORKDIR} 
   cp /home/users/${USER}/downloads/quantum_espresso_input_files/* ${WORKDIR}   cp /home/users/${USER}/downloads/quantum_espresso_input_files/* ${WORKDIR}
Line 70: Line 70:
 Bellow results show time of computation in function of used resources (computation scalability) for a specific computational task done with ''pw.x'' program. Bellow results show time of computation in function of used resources (computation scalability) for a specific computational task done with ''pw.x'' program.
  
-Assigning a larger amount of computing nodes does not always lead to a (efficient) reduction in computing time (wall-time of the job). To find the most appropriate number of nodes for a specific type of job, it is essential to run one's own benchmarks. In general, parallel jobs should scale to at least 70% efficiency for the sake of other TCR users. One user using twice the resources to squeeze out 10% more performance may be keeping other users from working at all. +Assigning a larger amount of computing nodes does not always lead to a (efficient) reduction in computing time (wall-time of the job). To find the most appropriate number of nodes for a specific type of job, it is essential to run one's own benchmarks. In general, parallel <color #22b14c>jobs should scale to at least 70% efficiency for the sake of other TCR users</color>. One user using twice the resources to squeeze out 10% more performance may be keeping other users from working at all.
- +
-To automate these tests 2 files were prepared ( ''run_qe-6.4.1_tests.sh'', ''qe-6.4.1.batch'' ). Test are run with script ''run_qe-6.4.1_tests.sh'' that prepares parameters and starts single 'batch' job (''qe-6.4.1.batch''). File ''qe-6.4.1.batch'' uses  [[https://slurm.schedmd.com/job_array.html|slurm's arrays]] that transform each submitted job to 3 separate computational jobs ( it is more efficient than submitting 3 separate jobs ).+
  
 Bellow results should be consider as results of this specific computational task on this specific hardware (TCR cluster) and not overall benchmark for Quantum Espresso software suite. Bellow results should be consider as results of this specific computational task on this specific hardware (TCR cluster) and not overall benchmark for Quantum Espresso software suite.
 +
 +{{ tcr:qe65_tcr_result_001.png?nolink|}}
 +^nodes ^min [s] ^avg [s] ^median [s] ^max [s] ^efficiency [%] ^
 +|1 |1862 |2065.75 |2126 |2149 |<color #22b14c>100.00%</color>|
 +|2 |1105 |1192.75 |1157.5 |1351 |<color #22b14c>84.25%</color>|
 +|3 |763 |770 |768.5 |780 |<color #22b14c>81.35%</color>|
 +|4 |798 |1026.25 |868.5 |1570 |58.33%|
 +|5 |571 |589.75 |574 |640 |65.22%|
 +|6 |488 |536 |525.5 |605 |63.59%|
 +|7 |372 |502.25 |414.5 |808 |<color #22b14c>71.51%</color>|
 +|8 |375 |489.75 |408.5 |767 |62.07%|
 +|9 |324 |403.75 |346.5 |598 |63.85%|
 +|10 |327 |456.25 |465.5 |567 |56.94%|
 +|12 |285 |429.4 |375 |609 |54.44%|
 +|16 |228 |274.25 |281.5 |306 |51.04%|
 +
 +
 +
 +
 +*) values (min, avg, median, max, efficiency) do not include failed runs \\
 +*) efficiency as ''t1 / ( nodes * tn )'' ( where t1 is min computation time at one node, tn is min computation time on N nodes ) \\
  
tcr_sri002.1587890010.txt.gz · Last modified: 2023/08/01 06:38 (external edit)

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki