The appended is a post I made to the WRF Users Forum on 08 June. The absence of replies there suggests nobody loves me on that forum, so I'll try another :)<br><br>Since the time of my post, I've also compiled this (using mpif90, etc.) on a Penguin Computing cluster of Opteron processors, and am running in the same problem. I've also removed the "PBS Script" interface and am simply using PBS to grab an interactive node, then running ./run_wrfpost straight from the command line. My questions are<br>
<br>1) Are any of you actually running wrfpost.exe in parallel?<br>2) Are there any "gotchas" I might want to be aware of before digging in deeper?<br><br>Thanks for any help,<br><br>Don Morton<br>Arctic Region Supercomputing Center<br clear="all">
<br>-- <br>Arctic Region Supercomputing Center<br><a href="http://www.arsc.edu/~morton/">http://www.arsc.edu/~morton/</a><br><br>============================================================<br><br>Howdy,<br><br>After a fair amount of compilation struggles, I managed to
compile the dmpar version of wrfpost.exe, and am now trying to run
wrfpost.exe on a Cray XT5 by inserting the following command line in
run_wrfpost:<br><br>aprun -n 8 ${POSTEXEC}/wrfpost.exe < itag >
wrfpost_${domain}.$fhr.out 2>&1<br><br>Then, I have run_wrfpost
called by a PBS script which allocates 8 cores. Although it does
execute, what I get for output looks something like:<br><br> <span style="color: rgb(255, 64, 64);"> we will try to run with 1
server groups<br> we will try to run with 1 server groups<br>
*** you specified 0 I/O servers <br> we will try to run with
1 server groups<br> we will try to run with 1 server
groups<br> CHKOUT will write a file<br> *** you specified 0 I/O
servers <br> *** you specified 0 I/O servers <br> CHKOUT will write a
file<br> CHKOUT will write a file<br> The Posting is using
1 MPI task<br> There are 0 I/O servers<br> The Posting
is using 1 MPI task<br> The Posting is using 1
MPI task<br> There are 0 I/O servers<br> There are
0 I/O servers<br> *** you specified 0 I/O servers <br> CHKOUT
will write a file<br> The Posting is using 1 MPI task<br>
There are 0 I/O servers<br> 0</span><br><br>So, the 8
tasks are launched but<br><br>a) Task 7 does not appear to take on the
role of an I/O server(the latest WRF-ARW user's guide seems to imply
that it should?)<br>b) It appears that each task is only aware of
itself, and not the other tasks. <br><br>The code actually runs, but
takes 9 minutes (1049x1049x51 gridpoints) whether I use 4 or 8 tasks.<br><br>There
are plenty of things I might be doing wrong, and I'm preparing to jump
into sorc/wrfpost/SETUP_SERVERS.f to start some tracing, but before I
get in too deep, I'm just wondering if anyone else out there has
experience in this area and is aware of any "gotchas" that might save me
a day or two!<br><br>I'm literate in MPI and such, so don't really need
a lesson in that aspect. If I have to, I'll try to figure out why the
call to mpi_comm_size() seems to be returning 1 for npes, rather than 8.<br><br>