Re: mpqc 2.3.1

[Lennart Sorensen <lsorense@csclub.uwaterloo.ca>]

On Fri, Jun 30, 2006 at 09:15:25AM +0200, Francesco Pietra wrote:
> Huge thanks to Lennart for putting deb packages of mpqc 2.3.1 on 
> 
> http://www.tinyplanet.ca/~lsorense/mpqc/ 
> 
> Also huge thanks to Curtis for a very complex MC Search OO input b3lyp that is 
> now running with Lennart's mpqc 2.3.1 on four nodes opteron amd64 for an 
> intriguing C22H25NO6 molecule. Contrary to the simple input (like for 
> water.in), where the calculation spopped before reaching convergence, here we 
> are at the 17th iteration and it continues. Hopefully to convergence, we will 
> see. There is an extremely long delay after each iteration now, that may 
> generate doubts whether the calculation halted or it is only matter of being 
> patient.

Well I hope it works.

> I realize that with my system I am asking too much to quantum mechanics, and I 
> am still far from the complexity of my actual work, where my aim is to 
> understand the basis of the processes that I observe spectrally.
> 
> Incidentally, is there a need to set the shared memory that mpqc can utilize 
> (I have a total of 8GB) or is that automatically recognized?

On a 64 bit system, you can use all your ram from any program, if it
wants to use it.  Nothing has to be done specially to make it happen.
Of course if the program doesn't want to use more ram, then it won't.

> Although too early for a report, I wished to acknowledge the effective help I 
> received from both the amd64 list and the mpqc list.

--
Len Sorensen