From owner-nwchem-users@emsl.pnl.gov Mon Aug 13 08:45:41 2007 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.13.8/8.13.8) with ESMTP id l7DFjejR026179 for ; Mon, 13 Aug 2007 08:45:41 -0700 (PDT) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.13.8/8.13.8/Submit) id l7DFjePj026178 for nwchem-users-outgoing-0915; Mon, 13 Aug 2007 08:45:40 -0700 (PDT) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-Ironport-SG: No_SBRS X-Ironport-SBRS: None X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Ao8CAGQZwEaAuwMF/2dsb2JhbACCbA X-IronPort-AV: E=Sophos;i="4.19,255,1183359600"; d="scan'208,217";a="28677057" In-Reply-To: <96f4bb620708120731q61823dc4rf9ee3fa0a2dbfb48@mail.gmail.com> References: <1186914765.46bee1cdd1f1f@webmail.stud.uni-graz.at> <96f4bb620708120731q61823dc4rf9ee3fa0a2dbfb48@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-24-406798280 Message-Id: Cc: NWChem Users From: Matthew Asplund Subject: Re: [NWCHEM] CCSD(T) file size Date: Mon, 13 Aug 2007 09:45:12 -0600 To: "Jeff Hammond" X-Mailer: Apple Mail (2.752.3) X-Scanned-By: MIMEDefang 2.62 on 192.168.200.10 X-Scanned-By: MIMEDefang 2.61 on 192.168.105.38 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.0.2 (www-ext.chem.byu.edu [192.168.200.10]); Mon, 13 Aug 2007 09:45:34 -0600 (MDT) Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk --Apple-Mail-24-406798280 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=ISO-8859-1; delsp=yes; format=flowed Is there an easy way to calculate how much memory memory a =20 calculation will take? I assume from the output that the job needs =20 1905911348 words 14 GB or memory (1905911348/1024/1024/1024*8 GB). =20 Is that the correct reading of things? Matthew Asplund On Aug 12, 2007, at 8:31 AM, Jeff Hammond wrote: > You need more memory. You can try reducing the tile size but that > isn't likely to help since you're way short on memory as best I can > tell. Are you running the job in parallel? With how much memory? > What are you memory settings in the input file? > > 400 basis functions with CCSD(T) is certainly doable with NWChem but > you might need quite a few processors depending on what symmetry your > molecule is. > > Please include the whole output file There's a lot of information in > it that can help us to better solve your problem. > > Jeff > > On 8/12/07, Walter M. F. Fabian wrote: >> >> Hi, >> I am trying to do a CCSD(T) calculation on a medium sized organic >> molecule (40 atoms, 400 basis functions) but get an error message of >> the type >> Is there any chance to do this calculation at all with NWChem; if =20= >> yes, >> what hardware requirements are recommended (with another software >> package I could do it; however on an older computer which will be =20= >> shut >> down by the end of the year and taking ca. 3 months. >> >> Any suggestions/recommendations are highly welcome >> >> >> Walter M.F. Fabian >> >> 4-electron integrals stored in orbital form >> 1: WARNING:armci_set_mem_offset: offset changed 1548288 to 1581056 >> -10004: WARNING:armci_set_mem_offset: offset changed 0 to 29634560 >> -10002: WARNING:armci_set_mem_offset: offset changed 0 to 29634560 >> -10006: WARNING:armci_set_mem_offset: offset changed 0 to 29634560 >> >> v2 file size =3D 1905911348 >> available GA memory 1763986312 bytes >> =20 >> ---------------------------------------------------------------------=20= >> --- >> createfile: failed ga_create size=3D********* >> =20 >> ---------------------------------------------------------------------=20= >> --- >> =20 >> ---------------------------------------------------------------------=20= >> --- >> current input line : >> 71: TASK TCE ENERGY >> =20 >> ---------------------------------------------------------------------=20= >> --- >> =20 >> ---------------------------------------------------------------------=20= >> --- >> =20 >> ---------------------------------------------------------------------=20= >> --- >> For more information see the NWChem manual at >> http://www.emsl.pnl.gov/docs/nwchem/nwchem.html >> >> >> For further details see manual section: >> >> >> >> >> available GA memory 1763986312 bytes >> available GA memory 1763986320 bytes >> available GA memory 1763986320 bytes >> createfile: failed ga_create size=3D********* >> =20 >> ---------------------------------------------------------------------=20= >> --- >> >> >> >> >> -- >> Dr. Walter M. F. Fabian >> Institut f=FCr Chemie, Karl-Franzens Universit=E4t Graz >> Heinrichstr. 28, A-8010 GRAZ, Austria >> e-mail: walter.fabian@uni-graz.at, www: http://www.uni-graz.at/=20 >> walter.fabian/ >> FAX: (austria) 316-380-9840, Tel.: (austria) 316-380-8636 >> >> ---------------------------------------------------- >> Sent through IMP on https://webmail.stud.uni-graz.at >> > > > --=20 > Jeff Hammond > The University of Chicago > Matthew Asplund Department of Chemistry and Biochemistry Brigham Young University Provo, UT asplund@chem.byu.edu (801) 422-5275 --Apple-Mail-24-406798280 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1 Is there an easy way to calculate how much memory memory a calculation = will take? =A0I assume from the output that the job needs=A01905911348 = words 14 GB or memory (1905911348/1024/1024/1024*8 GB). =A0Is that the = correct reading of things?

Matthew = Asplund

On Aug 12, 2007, at 8:31 AM, Jeff = Hammond wrote:

You need more memory.=A0 You can try reducing the tile = size but that
isn't likely to help since = you're way short on memory as best I can
=A0 Are you running = the job in parallel?=A0 = With how much memory?
What are you = memory settings in the input file?

400 basis functions with CCSD(T) = is certainly doable with NWChem but
you might = need quite a few processors depending on what symmetry your
molecule is.

Please include the whole output = file=A0 There's a lot of = information in
it that can help us to better = solve your problem.
Jeff

On = 8/12/07, Walter M. F. Fabian <walter.fabian@uni-graz.at>= ; wrote:

=A0Hi,
=A0I am = trying to do a CCSD(T) calculation on a medium sized organic
=A0molecu= le (40 atoms, 400 basis functions) but get an error message of
=A0the = type
=A0Is there any chance to do this = calculation at all with NWChem; if yes,
=A0what hardware requirements are = recommended (with another software
=A0package I could do it; however = on an older computer which will be shut
=A0down by the end of the year = and taking ca. 3 months.

=A0Any = suggestions/recommendations are highly welcome


=A0Walter M.F. Fabian

=A04-electron integrals stored in = orbital form
=A01: = WARNING:armci_set_mem_offset: offset changed 1548288 to = 1581056
=A0-10004: = WARNING:armci_set_mem_offset: offset changed 0 to 29634560
=A0-10002= : WARNING:armci_set_mem_offset: offset changed 0 to 29634560
=A0-10006= : WARNING:armci_set_mem_offset: offset changed 0 to 29634560

=A0 v2=A0 =A0 file size =A0 =3D =A0 =A0 =A0 1905911348
=A0=A0 = available GA memory =A0 =A0 = =A0 =A0 =A0 =A0 1763986312=A0= bytes
=A0 = -------------------------------------------------------------------= -----
=A0 createfile: failed ga_create = size=3D*********
=A0 = -------------------------------------------------------------------= -----
=A0 = -------------------------------------------------------------------= -----
=A0=A0 current input line = :
=A0=A0 =A0 = 71: TASK TCE ENERGY
=A0 = -------------------------------------------------------------------= -----
=A0 = -------------------------------------------------------------------= -----
=A0 = -------------------------------------------------------------------= -----
=A0 For more information see the = NWChem manual at


=A0 For = further details see manual section:




=A0=A0 available GA memory =A0 =A0 =A0 =A0 =A0 =A0 = 1763986312=A0 = bytes
=A0=A0 available GA memory =A0 =A0 =A0 =A0 =A0 =A0 = 1763986320=A0 = bytes
=A0=A0 available GA memory =A0 =A0 =A0 =A0 =A0 =A0 = 1763986320=A0 = bytes
=A0 createfile: failed ga_create = size=3D*********
=A0 = -------------------------------------------------------------------= -----




--
Dr. = Walter M. F. Fabian
Institut f=FCr Chemie, = Karl-Franzens Universit=E4t Graz
Heinrichstr. = 28,=A0 A-8010 GRAZ, = Austria
e-mail: = walter.fabian@uni-graz.at, www: http://www.uni-graz.at/walt= er.fabian/
FAX: (austria) 316-380-9840, = Tel.: (austria) 316-380-8636


=


--=A0
Jeff = Hammond
The University of = Chicago

=

Matthew = Asplund
Department of Chemistry and = Biochemistry=A0
Brigham Young University
Provo, = UT
(8= 01) 422-5275




 =

= --Apple-Mail-24-406798280--