software:lapack:caviness

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
software:lapack:caviness [2020-03-09 11:57] – [qsub file to test] anitasoftware:lapack:caviness [2021-04-27 16:21] (current) – external edit 127.0.0.1
Line 25: Line 25:
 already have.</note> already have.</note>
  
-===== Compiling with intel and mkl library =====+===== Compiling with Intel and MKL library =====
  
 The [[https://software.intel.com/en-us/parallel-studio-xe/|Intel Parallel Studio XE]] comes installed with a Fortran compiler with the MKL library.  Use VALET, **''vpkg_versions intel''**. to find the latest version installed on Caviness. The [[https://software.intel.com/en-us/parallel-studio-xe/|Intel Parallel Studio XE]] comes installed with a Fortran compiler with the MKL library.  Use VALET, **''vpkg_versions intel''**. to find the latest version installed on Caviness.
Line 45: Line 45:
  
 <code> <code>
 +workgroup -g <<investing-entity>>
 vpkg_devrequire intel vpkg_devrequire intel
 ifort -mkl dgels-ex.f -o dgels-ex ifort -mkl dgels-ex.f -o dgels-ex
Line 70: Line 71:
 </code> </code>
 </note> </note>
-==== qsub file to test ====+==== sbatch file to test ====
  
-The ''ifort'' compiler with flag ''-mkl'' will compile and link to the threaded MKL libraries.  Thus you should test in the threaded parallel environment, and export the number of slots to the ''MKL_NUM_THREAD'' environment variable. Remember to use our templates for threaded jobs which can be found in ''/opt/templates/slurm/generic/threads.qs'' as a starting point. Here is a simple ''test.qs'' based on the ''threads.qs'' template.+The ''ifort'' compiler with flag ''-mkl'' will compile and link to the threaded MKL libraries.  Thus you should test in the threaded parallel environment, and export the number of slots to the ''MKL_NUM_THREAD'' environment variable. Remember to use our templates for threaded jobs which can be found in ''/opt/shared/templates/slurm/generic/threads.qs'' as a starting point. Here is a simple ''test.qs'' based on the ''threads.qs'' template.
 <file bash test.qs> <file bash test.qs>
 #!/bin/bash -l #!/bin/bash -l
Line 162: Line 163:
 echo "" echo ""
 echo "--- Compare Results ---" echo "--- Compare Results ---"
-cat $JOB_NAME.r+cat $SLURM_JOB_NAME.r
 </file> </file>
  
 ==== Test result output ==== ==== Test result output ====
 <code> <code>
 +[traine@login01 nagex]$ workgroup -g it_css
 +[(it_css:traine)@login01 nagex]$ sbatch test.qs
 +Submitted batch job 6718859
 +[(it_css:traine)@login01 nagex]$ more slurm-6718859.out
 +-- OpenMP job setup complete:
 +--  OMP_THREAD_LIMIT     = 4
 +--  OMP_PROC_BIND        = true
 +--  OMP_PLACES           = cores
 +--  MP_BLIST             = 0,1,2,3
 +
 --- Set environment --- --- Set environment ---
-WARNING: 'gcc' was not found +Adding package `intel/2018u4` to your environment
-Adding package `intel/2013-2.144-64bit` to your environment+
  
 --- Run Test with 4 threads --- --- Run Test with 4 threads ---
  DGELS Example Program Results  DGELS Example Program Results
- +
  Least squares solution  Least squares solution
       1.5339     1.8707    -1.5241     0.0392       1.5339     1.8707    -1.5241     0.0392
- +
  Square root of the residual sum of squares  Square root of the residual sum of squares
       2.22E-02       2.22E-02
  
-real 0m0.966s +real    0m1.043s 
-user 0m0.003s +user    0m0.007s 
-sys 0m0.031s+sys     0m0.049s
  
 --- Compare Results --- --- Compare Results ---
Line 193: Line 203:
       2.22E-02       2.22E-02
 </code> </code>
- 
-<note warning>**WARNING: 'gcc' was not found**:  The release notes for this version of intel composer suite indicate 
-that GNU **''gdk''** is included for debugging, which requires **''gcc''**.  This warning can be ignored if you are not debugging with **''gdk''** in the batch script. You will not get this warning on the head node, since the system version of **''gcc''** will always be found in your path.  
- 
-If your are debugging on the compute nodes or want to remove the warning, add 
-<code> 
-vpkg_require gcc/4.6 
-</code> 
-before the intel ''vpkg_require'' command in you batch script file. 
-</note> 
  
 <note important>Sub-second timing results are not reliable.  This test is not a benchmark and was meant to show that <note important>Sub-second timing results are not reliable.  This test is not a benchmark and was meant to show that
Line 212: Line 212:
  
   * Programs with small arrays will not benefit from the multi-threaded library, and may suffer a bit from the system overhead of maintaining multiple threads.   * Programs with small arrays will not benefit from the multi-threaded library, and may suffer a bit from the system overhead of maintaining multiple threads.
-  * Sequential programs are better suited for running simultaneous instances.  You could run 12 copies of the program on the same node with better throughput when you compile them to be sequential.  (Too many threads on the same node will contend for limited resources)+  * Sequential programs are better suited for running simultaneous instances.  You could run ''n'' copies of the program on the same node, where ''n'' is the number of cores on that node, with better throughput when you compile them to be sequential.  (Too many threads on the same node will contend for limited resources)
   * You may be able to take control of the parallelism in your program with OPENMP compiler directions.  This is easiest if you using the single threaded MKL in your parallel regions. See [[https://software.intel.com/en-us/articles/recommended-settings-for-calling-intelr-mkl-routines-from-multi-threaded-applications|recommended settings for calling intel MKL routines from multi threaded applications]].   * You may be able to take control of the parallelism in your program with OPENMP compiler directions.  This is easiest if you using the single threaded MKL in your parallel regions. See [[https://software.intel.com/en-us/articles/recommended-settings-for-calling-intelr-mkl-routines-from-multi-threaded-applications|recommended settings for calling intel MKL routines from multi threaded applications]].
  
-===== Compiling with PGI and ACML library ===== 
- 
-The [[http://developer.amd.com/tools-and-sdks/cpu-development/amd-core-math-library-acml//|AMD core math library (ACML)]] is from AMD developers, and is thus a good chioce form the Mills chip set. Use VALET, **''vpkg_versions acml''**. to find the latest version installed on Mills - ''5.3.0''. 
- 
-<note tip>**Versions:** 
-From the release notes in the file ''/opt/shared/ACML/5.3.0/ReleaseNotes'' 
-   New features of release 5.3.0 of ACML 
-     Updated the LAPACK code to version 3.4.0. 
-</note> 
  
          
  • software/lapack/caviness.1583769422.txt.gz
  • Last modified: 2020-03-09 11:57
  • by anita