[molpro-user] Molpro hangs with RASSCF with large active space
KnowlesPJ at Cardiff.ac.uk
Tue Mar 1 08:48:22 GMT 2011
It wouldn't be a surprise to me if this job happened to need a very long time to complete. The implementation of 'restrict' in Molpro's MCSCF is functional but not necessarily optimal, and so hopeful comparison against other software options might not be the best way to estimate feasibility.
On 25 Feb 2011, at 22:56, Leon Freitag wrote:
> Dear Molpro users,
> I'm trying to run a large almost full valence + 9 Rydberg orbitals RASSCF
> calculation with fluoroethylene -- RAS (16,23) with 14 orbitals in RAS2 and 9
> orbitals in RAS3), and Molpro 2010 just gets stuck eating CPU time.
> Now that wouldn't be that surprising for such a large calculation, but it
> seems to me, that it gets stuck even before the first iteration. There's also
> no hard drive I/O, and memory usage does not increase. The compute node I'm
> trying to run this on has 4 quad core Intel Xeons E7330 with 128 GB of RAM, so
> at least hardware wise the system should be capable to run this.
> I also ran the same calculation in Molcas on the same machine; it finished the
> first RASSCF iteration in about two weeks (using just one CPU), but as Molcas'
> RASSCF code isn't parallel, I stopped the Molcas job in favour of Molpro using
> all 16 cores. Now Molpro is stuck for another two weeks hogging 16 cores, but
> has not produced any signs of even entering the first iteration.
> Input and output (so far) are attached.
> Does anyone have a clue what is going on there?
> Thanks in advance,
> P.S. A note for the moderator: I was posting this from the wrong adress
> Molpro-user mailing list
> Molpro-user at molpro.net
Prof. Peter J. Knowles
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff CF10 3AT, UK
Email KnowlesPJ at Cardiff.ac.uk
Secretary: Alison Thompson, ThompsonAF at Cardiff.ac.uk Telephone +44 29 208 74805
More information about the Molpro-user