AMBER Archive (2000)

Subject: Re: nmode out of memory

From: David Case (case_at_scripps.edu)
Date: Thu Sep 28 2000 - 10:09:53 CDT

On Thu, Sep 28, 2000, schayg_at_puskin-a206.sote.hu wrote:

> I am trying a normal mode calculation on a protein with 320 residues.
>
> Total memory required : 129902965 real words

In this case, a "real word" holds a double precision varaible = 8 bytes.
Hence, you need a little over 1 Gbyte of memory to run this job.

Here's why: 320 residues is probably something like 5000 atoms, which means
3*5000 = 15000 x,y,z coordinates. To hold the second derivative matrix,
you would need 15000*15000/2 = 112000000 double precision numbers, or
about 1 Gb of memory just for that part.

> i am trying to run this on a Linux box with 64M memory and 130M virtual

This is unfortunately not really a feasible calculation for this platform.
Very different algorithms (probably taking advantage of sparse matrix
techniques) would be required for such a calculation.

>
> i am also wondering about the available memory it reported, it seems very
> low...

This is set in sizes.h, which has more explanation about these things.

...hope this helps...dac 

-- 

==================================================================
David A. Case                     |  e-mail:      case_at_scripps.edu
Dept. of Molecular Biology, TPC15 |  fax:          +1-858-784-8896
The Scripps Research Institute    |  phone:        +1-858-784-9768
10550 N. Torrey Pines Rd.         |  home page:                   
La Jolla CA 92037  USA            |    http://www.scripps.edu/case
==================================================================