===== Performance Reports =====
 

==== Synopsis: ==== 
''perf-report'' is a lightweight profiling tool that provides basic information about where the computing time is actually spent in a given application. It is developed by ARM (formerly Allinea) [[https://developer.arm.com/tools-and-software/server-and-hpc/debug-and-profile/arm-forge]] and very easy to use, i.e. by simply prefixing the usual executable with a call to ''perf-report''.
It is nowadays integral part of [[doku:forge|ARM's forge]].

==== Usage: ====
For example, analyzing a simple MPI job could be done with the help of the following submit script to [[doku:slurm|SLURM]]

   #!/bin/bash
   #
   #SBATCH -J prflng        
   #SBATCH -N 2 
   #SBATCH --ntasks-per-node=16
   #SBATCH --ntasks-per-core=1
  
   module purge
   module load openmpi/4.1.4-gcc-12.2.0-x6sxxj3 arm/20.1_FORGE
    
   perf-report srun --jobid $SLURM_JOB_ID --mpi=pmi2 -n 32 ./a.out
   
This will result in the creation of two summary files in *.txt and *.html format providing an overview of the relative time spent in MPI, I/O, OpenMP etc. Note the 'Energy' section in recent releases !

==== Further Reading: ====
''/opt/sw/x86_64/glibc-2.17/ivybridge-ep/allinea/20.1_FORGE/doc/userguide-forge.pdf'' ( part 4, explanations, examples, typical cases  )