From owner-chemistry@ccl.net Fri Dec 31 13:06:00 2010 From: "Jean Jules FIFEN julesfifen===gmail.com" To: CCL Subject: CCL:G: CASSCF Memory Requirements Message-Id: <-43484-101231110420-2910-FePb7F7dQfjC0zgODhJWkw^server.ccl.net> X-Original-From: Jean Jules FIFEN Content-Type: multipart/alternative; boundary=90e6ba53a9f4d365620498b6f31a Date: Fri, 31 Dec 2010 17:04:08 +0100 MIME-Version: 1.0 Sent to CCL by: Jean Jules FIFEN [julesfifen(a)gmail.com] --90e6ba53a9f4d365620498b6f31a Content-Type: text/plain; charset=ISO-8859-1 Dear Soren Eustis, Reading the error message, I think you have not provided sufficient memory required by the job. The job required at least 33GB of RAM memory. So, specified it by %mem=33GB. If you do not have sufficient physical memory, you should lower the level of computation either by reducing the number of basis functions or by freezing more orbitals in CASSCF procedure. Best regards, On 31 December 2010 10:11, Soren Eustis soren]~[env.ethz.ch < owner-chemistry_+_ccl.net> wrote: > > Sent to CCL by: "Soren Eustis" [soren+*+env.ethz.ch] > Greetings. I am trying to do a CASSCF(6,6) conical optimization of a > medium sized molecule. Using the 6-31++G basis set, I end up with 350 basis > functions. This is y first foray into CASSCF on Gaussian, and I feel that I > must be making an error because the job always halts with a not enough > memory error. I have tried increasing up to 8gb per processor (!) running 4 > processors. This cannot be correct! I was being frugal by choosing a (6,6) > active space, yet I am still running out of resources. I was unable to find > a good discussion of casscf memory issues, so I am posting my question here. > I would greatly appreciate some help with this. Abridged input and output > below. > > Regards, > > Soren Eustis > ETH - Zurich > > ************************************* > Gaussian 09: AM64L-G09RevA.02 11-Jun-2009 > 31-Dec-2010 > ****************************************** > %chk=CASSCFConical.chk > %mem=32GB > %nprocshared=8 > Will use up to 8 processors via shared memory. > Default route: MaxDisk=2GB > ---------------------------------------------------------------------- > # opt(conical) casscf(6,6,slaterdet,nroot=2)/6-31++g scrf=(iefpcm,solv > ent=water) guess=save pop=savenlmo geom=connectivity > > > **** > > electronic state of the initial guess is 1-A. > Out-of-memory error in routine OMatEl (IEnd= 4294966401 MxCore= > 4294966296) > Use %mem=4097MW to provide the minimum amount of memory required to > complete this step. > Error termination via Lnk1e in /cluster/apps/gaussian/g09/l405.exe at Fri > Dec 31 06:07:36 2010. > Job cpu time: 0 days 0 hours 0 minutes 19.1 seconds. > File lengths (MBytes): RWF= 89 Int= 0 D2E= 0 Chk= 1 > Scr= 1> > > -- J. Jules. --90e6ba53a9f4d365620498b6f31a Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Dear Soren Eustis,
Reading the error message, I think y= ou have not provided sufficient memory required by the job. The job require= d at least 33GB of RAM memory. So, specified it by %mem=3D33GB. If you do n= ot have sufficient physical memory, you should lower the level of computati= on either by reducing the number of basis functions or by freezing more orb= itals in CASSCF procedure.

Best regards,

On 31 December 2010 10:= 11, Soren Eustis soren]~[env.ethz.ch <owner-chemist= ry_+_ccl.net> wrote:

Sent to CCL by: "Soren =A0Eustis" [soren+*+env.ethz.ch]
Greetings. =A0I am trying to do a CASSCF(6,6) conical optimization of a med= ium sized molecule. =A0Using the 6-31++G basis set, I end up with 350 basis= functions. =A0This is y first foray into CASSCF on Gaussian, and I feel th= at I must be making an error because the job always halts with a not enough= memory error. =A0I have tried increasing up to 8gb per processor (!) runni= ng 4 processors. =A0This cannot be correct! =A0I was being frugal by choosi= ng a (6,6) active space, yet I am still running out of resources. =A0I was = unable to find a good discussion of casscf memory issues, so I am posting m= y question here. I would greatly appreciate some help with this. =A0Abridge= d input and output below.

Regards,

Soren Eustis
ETH - Zurich

*************************************
=A0Gaussian 09: =A0AM64L-G09RevA.02 11-Jun-2009
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A031-Dec-2010
=A0******************************************
=A0%chk=3DCASSCFConical.chk
=A0%mem=3D32GB
=A0%nprocshared=3D8
=A0Will use up to =A0 =A08 processors via shared memory.
=A0Default route: =A0MaxDisk=3D2GB
=A0---------------------------------------------------------------------- =A0# opt(conical) casscf(6,6,slaterdet,nroot=3D2)/6-31++g scrf=3D(iefpcm,so= lv
=A0ent=3Dwater) guess=3Dsave pop=3Dsavenlmo geom=3Dconnectivity


****

electronic state of the initial guess is 1-A.
=A0Out-of-memory error in routine OMatEl (IEnd=3D =A0 =A04294966401 MxCore= =3D =A0 =A04294966296)
=A0Use %mem=3D4097MW to provide the minimum amount of memory required to co= mplete this step.
=A0Error termination via Lnk1e in /cluster/apps/gaussian/g09/l405.exe at Fr= i Dec 31 06:07:36 2010.
=A0Job cpu time: =A00 days =A00 hours =A00 minutes 19.1 seconds.
=A0File lengths (MBytes): =A0RWF=3D =A0 =A0 89 Int=3D =A0 =A0 =A00 D2E=3D = =A0 =A0 =A00 Chk=3D =A0 =A0 =A01 Scr=3D =A0 =A0 =A01



-=3D This is automatically added to each message by the mailing script =3D-=
E-mail to subscribers: CHEMISTRY_+_ccl.n= et or use:
=A0 =A0 =A0http://www.ccl.net/cgi-bin/ccl/send_ccl_message

E-mail to administrators: CHEM= ISTRY-REQUEST_+_ccl.net or use
=A0 =A0 =A0http://www.ccl.net/cgi-bin/ccl/send_ccl_message

Subscribe/Unsubscribe:
=A0 =A0 =A0http://www.ccl.net/chemistry/sub_unsub.shtml

Before posting, check wait time at: http://www.ccl.net

Job: http://www.ccl.n= et/jobs
Conferences: http://server.ccl.net/chemistry/announcements/co= nferences/

Search Messages: http://www.ccl.net/chemistry/searchccl/index.shtml
=A0 =A0 =A0h= ttp://www.ccl.net/spammers.txt

RTFI: http://www.ccl.net/chemistry/aboutccl/instructions/





--
J. Jul= es.

--90e6ba53a9f4d365620498b6f31a--