[molpro-user] caspt3 error

Manhui Wang wangm9 at cardiff.ac.uk
Mon Jan 31 13:15:14 GMT 2011


Dear Benjamin,

I am wondering whether there is a problem in caspt3 or your
installation. Have you validated your installation(i,e, all the testjobs
pass)?  We have seen some problems for Molpro with openmpi in cases
large memory is used or  jobs are run on multinodes. If this is your
case, can you try mvapich2 1.5.1p1, which doesn't suffer such problem.

Best wishes,
Manhui


Mintz, Benjamin J. wrote:
> Users,
> 
> The previous email came from Ben Mintz.  My signature did not get attached.
> 
> Benjamin J. Mintz
> Research Associate, Postdoctoral
> Oak Ridge Leadership Computing Facility (OLCF)
> Oak Ridge National Laboratory
> Phone: (865) 574-6639
> Email: mintzbj at ornl.gov
> Gmail: bm0037 at gmail.com
> 
>>> To Users,
>>>
>>> I am trying to run a caspt3 computation on the 5th excited state of a radical.  I am getting the following in my output, which is given from a parallel molpro mppx version that was compiled using global-arrays and openmpi 1.4.3 and an infiniband network. This output came from a single process job. I was getting similar results for multiprocess jobs.  I requested 400 mw of memory for this job. I also tried requesting 3000 mw just to test if this was a memory issue, and I got the same results.  The compute nodes have 2 quad-core intel nehalem processes with 32 GB of RAM.  Why am I getting a negative value for the available memory?  Any advice would be appreciated. 
>>>
>>>
>>> ARCHNAME  : Linux/x86_64
>>> FC        : /cerebro/home/software/intel/composerxe-2011.0.084/bin/intel64/ifort
>>> FCVERSION : 12.0.0
>>> BLASLIB   : /cerebro/home/software/intel/mkl/lib/intel64/libmkl_solver_ilp64_sequential.a -Wl,--start-group /cerebro/hom
>>> e/software/intel/mkl/lib/intel64/libmkl_intel_ilp64.a /cerebro/home/software/intel/mkl/lib/intel64/libmkl_sequential.a /c
>>> erebro/home/software/intel/mkl/lib/intel64/libmkl_core.a -Wl,--end-group -lpthread
>>> id        : vt
>>>
>>> Nodes             nprocs
>>> compute-0-10.local    1
>>> ga_uses_ma=false, calling ma_init with nominal heap.
>>> GA-space will be limited to   8.0 MW (determined by -G option)
>>>
>>> .
>>> .
>>> .
>>>
>>> Reference energy                    -192.02654945
>>> Nuclear energy                       128.02067437
>>> Kinetic energy                       192.18101184
>>> One electron energy                 -510.85830694
>>> Two electron energy                  190.24161912
>>> Virial quotient                       -1.00215943
>>> Correlation energy                    -0.56946400
>>> !RSPT2 STATE 5.1 Energy             -192.596013448829
>>>
>>> !RSPT2 STATE 5.1 Dipole moment         0.00000000     0.00000000     0.55143007
>>> Dipole moment /Debye                   0.00000000     0.00000000     1.40150364
>>>
>>> !RSPT expec            <5.1|H|5.1>  -192.500573879556
>>>
>>> Correlation energy                    -0.58781752
>>> !RSPT3 STATE 1.1 Energy             -192.614366961542
>>> insufficient memory available - require                      0  have 
>>> -2314885530418453536
>>> the request was for real words
>>>
>>>
>>>
>>
>>
>>
>>
> 
> 
> 
> 
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user

-- 
-----------
Manhui  Wang
School of Chemistry, Cardiff University,
Main Building, Park Place,
Cardiff CF10 3AT, UK



More information about the Molpro-user mailing list