"NOT ENOUGH CORE SPACE FOR SORTING IN TRA2" problem when performing geometry optimization with RASSCF

Posted by Evgeniy on January 27, 2009 at 11:18:38:

Dear Developers and Users of Molcas,

I encountered a problem when attempting to perform geometry optimization at the CASSCF level for an excited state with analytical gradients. The rasscf module ran fine, but the mclr module crashes with the diagnostic:

NOT ENOUGH CORE SPACE FOR SORTING IN TRA2
TOTAL SORTING SPACE IS 1572535272
STEP1: AVAILABLE IS 844945920 NEEDED IS 39060
STEP2: " 844945920 " 39060
STEP3: " 0 " 503360
STEP4: " 844945920 " 72540

I set 12000 MB in MOLCASMEM. Is it still not enough?
The system is pretty large - 28 atoms, 279 MOs no symmetry. Any idea to sort it out would be greately appreciated.

A related question: are *analytical* gradients available for some method (SCF, CASSCF) when one uses Cholesky decomposition? Thanks!

Best,
Evgeniy

Follow Ups:

Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:
: Dear Developers and Users of Molcas, : I encountered a problem when attempting to perform geometry optimization at the CASSCF level for an excited state with analytical gradients. The rasscf module ran fine, but the mclr module crashes with the diagnostic: : NOT ENOUGH CORE SPACE FOR SORTING IN TRA2 : TOTAL SORTING SPACE IS 1572535272 : STEP1: AVAILABLE IS 844945920 NEEDED IS 39060 : STEP2: " 844945920 " 39060 : STEP3: " 0 " 503360 : STEP4: " 844945920 " 72540 : I set 12000 MB in MOLCASMEM. Is it still not enough? : The system is pretty large - 28 atoms, 279 MOs no symmetry. Any idea to sort it out would be greately appreciated. : A related question: are *analytical* gradients available for some method (SCF, CASSCF) when one uses Cholesky decomposition? Thanks! : Best, : Evgeniy

[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]