[molpro-user] questions on -d option and RCCSD(T)-F12

[Falcetta, Michael F.]

Hi,

I am running a molpro calculation on a radical anion and want to do a CCSD(T)-F12 calculation and it is turning out to be a bit large using the UCCSD(T)-F12 option.  I was wondering whether the RCCSD(T)-F12 calculation would reduce the scratch file demands.

Also, The system on which I am running is geared to use as much RAM as possible for the scratch directory  ( NOT the memory for the calculation, but using a RAMDISK for the scratch area to speed up I/O), but the calculation gets too big.  If I specify two directories with the -d option in the molpro call, will the program automatically switch to the second directory once it uses all the available space on the ramdisk - or is there a way to tell the program how large it should allow anything to grow on that ramdisk?

Basically the computer tells me I am out of memory, but not because of the actual memory use by the -m option, but because the remaining memory, being used for the ramdisk scratch area gets too large.

Any help on either count is appreciated.

Mike

Michael F. Falcetta, Ph. D.
Professor, Chemistry Department
Box 3104, Grove City College
Grove City PA 16127
724.458.3820 (office)
724.372.0098 (mobile)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20131016/ef911ac1/attachment.html>