Re: CCL:Gaussian03 Disk Space Error in MP2

Hi Cherif,
 you can try and split the .rwf files in smaller that 2 GB increments, 1.9 is
 good.
 As well, try to monitor your .rwf in the scratch directory during calculation to
 see at what point the jobs is terminated.
 Regards,
 Mima Staikova
 Quoting Cherif Matta <Cherif.Matta!at!Dal.Ca>:
 > Dear all,
 >
 > We are trying to perform a single point calculation on a fairly large
 > system
 > using G03. We have a disk-space error in spite of the availability of much
 > more
 > free disk space than required in the error message.
 >
 > This is a neutral closed-shell system with 1014 basis functions, 1548
 > primitive
 > gaussians, and 272 electrons.  We are running it on a 64-bit SuSe9.1 Linux
 > Box
 > with 100 GB empty disk space.  Since this is not a 32-bits system, the file
 > size should not be a problem.
 >
 > The following is the Gaussian03 input line:
 > ============================================
 > #p MP2(full, verystingy)/6-311++g** output=wfn density=current
 > MaxDisk=65GB scf=(MaxCycle=200, tight) 6d 10f
 >
 > After the convergence of the SCF, the job is aborted, and
 > the following is the error we get:
 > ==================================
 > Cycle   2  Pass 1  IDiag  1:
 >  RMSU=  4.56D-08    CP:  1.00D+00
 >  E= -1864.55241471435     Delta-E=       -0.000000000020 Rises=F Damp=F
 >  DIIS: error= 1.72D-09 at cycle   2 NSaved=   2.
 >  NSaved= 2 IEnMin= 2 EnMin= -1864.55241471435     IErMin= 2 ErrMin=
 > 1.72D-09
 >  ErrMax= 1.72D-09 EMaxC= 1.00D-01 BMatC= 9.11D-16 BMatP= 1.96D-14
 >  IDIUse=1 WtCom= 1.00D+00 WtEn= 0.00D+00
 >  Coeff-Com:  0.195D-01 0.980D+00
 >  Coeff:      0.195D-01 0.980D+00
 >  Gap=     0.295 Goal=   None    Shift=    0.000
 >  RMSDP=6.47D-09 MaxDP=6.88D-07 DE=-2.00D-11 OVMax= 5.82D-09
 >
 >  SCF Done:  E(RHF) =  -1864.55241471     A.U. after    2 cycles
 >              Convg  =    0.6466D-08             -V/T =  2.0011
 >              S**2   =   0.0000
 >  KE= 1.862590819875D+03 PE=-1.379542208284D+04 EE= 5.357304786856D+03
 >  Leave Link  502 at Sat Apr  2 04:00:29 2005, MaxMem=   31457280 cpu:
 > 11736.6
 >  (Enter /home/programs/g03/l801.exe)
 >  Range of M.O.s used for correlation:     1  1009
 >  NBasis=  1014 NAE=   136 NBE=   136 NFC=     0 NFV=     0
 >  NROrb=   1009 NOA=   136 NOB=   136 NVA=   873 NVB=   873
 >
 >  **** Warning!!: The largest alpha MO coefficient is  0.13325828D+03
 >
 >  Leave Link  801 at Sat Apr  2 04:00:29 2005, MaxMem=   31457280 cpu:
 > 0.1
 >  (Enter /home/programs/g03/l906.exe)
 >  Not enough disk for semidirect, minimum=   946930225
 >  Not enough memory for fully direct, minimum=   141181640.
 >  Not enough resources for E2 calculation.
 >  Error termination via Lnk1e in /home/programs/g03/l906.exe at Sat Apr  2
 > 04:00:30 2005.
 > ========================================================
 >
 > We tried to run the job on a cluster with RedHat Linux where the scratch
 > files
 > were broken, each into 6 x 2GB-files, but we got exactly the same error
 > message
 > despite the availability of a huge disk space.
 >
 > Can someone help with this? I will summarize.
 >
 > Thank you very much in advance.
 >
 > Cheers,
 >
 > Cherif
 >
 > -------------------------------------------------------
 > Cherif F. Matta, Ph.D., MCIC
 > * Izaak Walton Killam Post Doctoral Fellow
 >   Department of Chemistry, Dalhousie University
 >   Halifax, NS, Canada.
 > * Adjunct Professor of Chemistry
 >   McMaster University, Hamilton, ON, Canada.
 > -------------------------------------------------------
 > * Web site:  http://chem.utoronto.ca/~cmatta/
 >   Tel: (902) 494 7021      Fax: (902) 494 1310
 > -------------------------------------------------------
 >
 >
 >
 > -= This is automatically added to each message by the mailing script =-
 > To send e-mail to subscribers of CCL put the string CCL: on your Subject:
 > line
 > and send your message to:  CHEMISTRY!at!ccl.net
 >
 > Send your subscription/unsubscription requests to:
 CHEMISTRY-REQUEST!at!ccl.net
 >
 > HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs
 >
 > If your is mail bouncing from ccl.net domain due to spam filters, please
 > use the Web based form from CCL Home Page
 > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 >
 >
 >
 >
 >
 --
 Mima G. Staikova Ph.D.
 Department of Chemistry
 Undergraduate Computer Lab
 80 St. George Street
 University of Toronto
 Toronto, Ontario M5S 3H6
 phone: (416) 978-1388
 email: mstaikov!at!chem.utoronto.ca
 -------------------------------------------------
 This mail sent through IMP: http://horde.org/imp/