<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p>Dear colleagues, <br>
</p>
<p>I'm Writing to our community <span id="result_box"
class="short_text" lang="en"><span>to</span> <span>clarify</span>
<span>once and for all issues with parallel pwscf execution. <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>By
lecture of PWSCF manual, I know that there are several levels
of parallelization, including FFTW and more. At the same time,
I'm quite sure that most of beginner QE users do not have
complicated clusters, but are rather starting with decent
multicore PC's. <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>There
are two main parallel mechanisms (known to me) namely: MPI
and OpenMP. Because I'm using Ubuntu, there is free Hydra MPI
software (mpich) and also free OpenMP. From what I know
(please correct me if this is wrong) MPI software is used for
parallelization within one multicore PC and OpenMP is used to
make parallelization between several PC's. <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>So my
reasoning is: because I have 2 multicore PCs (namely PC1: 6
cores 2 threads, and PC2: 4 cores 2 threads, giving in overall
PC1: 12 and PC2 8 'cores') on both PC1 and PC2 I've installed
hydra MPI (for parallelization within multicores) and on both
there are OpenMP for communication between PC1 and PC2 to act
as 'cluster'. <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>I've
configured OpenMP so both machines sees one another and when
I'm starting some pw.x job using openmp on both machines i see
that processors are 100% busy. Problem is that in produced
output files I can see several repeats of pwscf output (just
like several separate pwscf processes write into one file at
the same time). <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>The QE
was configured and compiled with --enable-openmp on both
machines and when I simply run pw.x on PC1 it writes: <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span> "Parallel
version (MPI & OpenMP), running on 12 processor
cores"</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>I'm
running job with this command: <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span> mpirun.openmp
-np 20 pw.x -in job.in > job.out</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span> and as
said before, both machines are occupied with this job (tested
using htop).</span></span></p>
<p><span tabindex="-1" id="result_box" class="" lang="en"><span>Looking
for help</span> <span>in this topic I often</span> <span>get</span>
<span class="">to the truth</span> <span>'</span><span>please
contact</span> <span>your system administrator</span><span>',
but sadly I am the administrator! <br>
</span></span></p>
<p><span tabindex="-1" id="result_box" class="" lang="en"><span>Could
some one with a similar hardware configuration could comment
here how to achieve properly working cluster? <br>
</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>
Sincerely yours</span></span></p>
<p><span id="result_box" class="short_text" lang="en"><span>Konrad
Gruszka<br>
</span></span></p>
</body>
</html>