[Columbus] MCSCF calculation (single-point)
Thomas Mueller
th.mueller at fz-juelich.de
Mon Nov 6 04:59:18 EST 2017
On Wed, 1 Nov 2017, Filipe Belarmino wrote:
Dear Filipe,
you are running 464 basis functions without symmetry -
this requires about (((464*465)/2)**2 )/131072 MB
core memory (in-core version) and about 2 times this
size disk space for the AO-MO trafo step -
sort of 100 GB memory and 200 GB disk.
Keep the files preferentially on a SSD to avoid excessive
I/O wait times.
Regards,
Thomas
>
> Dear developers and users,
>
>
>
> I got the following error:
>
>
> "mcscfls" file
> --------------------------------------------------------------------------------
> --//------------------------------------------------//--------------------------
> ---------
> *** Starting integral transformation ***
>
> module tranlib input parameters:
>
> prnopt = 1, chkopt = 0,ortopt = 0, denopt = 0
> mapin(1 ) = 1, nsymao = 1, naopsy(1) = 464, freeze(1) = 1
> mapout(1) = 1, nsymmo = -1, nmopsy(1) = -1, fsplit = 1
> outlab = 0, seward = 0, lumorb = 0, DALTON2 = 0
> nextint = 2
> LDAMIN = 127, LDAMAX =107711, LDAINC = 64
> LRC1MX = -1, LRC2MX = -1, LRCSCR = 65000
>
> THRESH = 5.0000E-12 [cutoff threshold]
>
> module tranlib: workspace lcore=1309673494
>
> inoutp: segmentation information:
> in-core transformation space, avcinc =1308864214
> address segment size, sizesg =1308538520
> number of in-core blocks, nincbk = 0
> number of out-of-core blocks, noutbk = 1
> number of in-core segments, incseg = 0
> number of out-of-core segments, outseg = 9
> - Transformation proceeds out of core!
> *** If possible, increase the amount of core memory. ***
> bummer (fatal):writda: i/o error, (ioerr*100+unit)=2533
> "mcscfls" 229L, 8477C
> --------------------------------------------------------------------------------
> --//------------------------------------------------//--------------------------
> ---------
>
>
> "runc.error" file
>
>
> end of hernew
> IGNORING BNDCHK
> *** End of DALTON calculation ***
> end of mcdrt
> end of mcuft
> # records (trda3s)= 118668
> predicted trda3s size: ********** 118668
> calling trouta
> total number of buckets = 3
> Reading bucket # 1 1 1
> forrtl: No space left on device
> forrtl: severe (38): error during write, unit 99, file
> /state/partition1/fbelarmino/34154.lmmrq.quimica.ufpb.br/WORK/bummer
> Image PC Routine Line Source
> mcscf.x 000000000058A299 Unknown Unknown Unknown
> mcscf.x 00000000005F40CE Unknown Unknown Unknown
> mcscf.x 00000000005EF307 Unknown Unknown Unknown
> mcscf.x 000000000056FD06 Unknown Unknown Unknown
> mcscf.x 0000000000560688 Unknown Unknown Unknown
> mcscf.x 0000000000433D77 Unknown Unknown Unknown
> mcscf.x 0000000000431C6F Unknown Unknown Unknown
> mcscf.x 00000000004300B9 Unknown Unknown Unknown
> mcscf.x 000000000040F518 Unknown Unknown Unknown
> mcscf.x 0000000000407787 Unknown Unknown Unknown
> mcscf.x 000000000045C864 Unknown Unknown Unknown
> mcscf.x 0000000000454427 Unknown Unknown Unknown
> mcscf.x 000000000044D8D3 Unknown Unknown Unknown
> mcscf.x 00000000004042DE Unknown Unknown Unknown
> libc.so.6 0000003BAEE1ED1D Unknown Unknown Unknown
> mcscf.x 00000000004041E9 Unknown Unknown Unknown
>
> --------------------------------------------------------------------------------
> --//------------------------------------------------//--------------------------
> ---------
>
>
> "runls" file
>
>
> Nuclear repulsion energy: 315.6575489177
>
> Irred. repr. : A
> # orbitals : 464
> ---------------------------------------------------
> >>>>>>>>>>>>> info: preparing drt no 1 out of 1 <<<<<<
> starting mcdrt.x -m 40000 < mcdrtin > mcdrtls at Wed Nov 1 11:19:37 BRT 2017
>
> -------------------- mcdrt.x ------------------------
> Spin multiplicity: singlet
> Spatial symmetry (#): 1
> Number of Configurations: 20
> ---------------------------------------------------
> starting mcuft.x at Wed Nov 1 11:19:37 BRT 2017
>
> -------------------- mcuft.x ------------------------
> ---------------------------------------------------
> ++++++++++++++++++++ MCSCF calculation +++++++++++++++
>
> info: taking mocoef file from
> /state/partition1/fbelarmino/34154.lmmrq.quimica.ufpb.br/mocoef
> starting mcscf.x -m 40000 at Wed Nov 1 11:19:38 BRT 2017
>
> current reclen= 107711
> opened reclen= 107711
> recnum= 53392
>
> *** Error occured! ***
>
> mcscf.x 0000000000431C6F Unknown Unknown Unknown
> mcscf.x 00000000004300B9 Unknown Unknown Unknown
> mcscf.x 000000000040F518 Unknown Unknown Unknown
> mcscf.x 0000000000407787 Unknown Unknown Unknown
> mcscf.x 000000000045C864 Unknown Unknown Unknown
> mcscf.x 0000000000454427 Unknown Unknown Unknown
> mcscf.x 000000000044D8D3 Unknown Unknown Unknown
> mcscf.x 00000000004042DE Unknown Unknown Unknown
> libc.so.6 0000003BAEE1ED1D Unknown Unknown Unknown
> mcscf.x 00000000004041E9 Unknown Unknown Unknown
>
> --------------------------------------------------------------------------------
> --//------------------------------------------------//--------------------------
> ---------
>
> How can I fix it?
>
>
>
>
-----------------------------------------------------------
Dr. Thomas Mueller
Institute for Advanced Simulation (IAS)
Juelich Supercomputing Centre (JSC)
Phone: +49-2461-61-3175
Fax: +49-2461-61-6656
E-mail: th.mueller at fz-juelich.de
WWW: http://www.fz-juelich.de/jsc/
JSC is the coordinator of the
John von Neumann Institute for Computing (NIC)
and member of the
Gauss Centre for Supercomputing (GCS)
-----------------------------------------------------------
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
More information about the Columbus
mailing list