AMBER Archive (2005)

Subject: Re: AMBER: RE:Amber 8 with lam compilation problem

From: Amber admin (amber-admin_at_scripps.edu)
Date: Wed Feb 23 2005 - 14:39:02 CST

(From Bob Duke:)

Marciej -
Sorry, I have not been following this thread.

There are instructions for config files for pmemd when using ifort 8 and
mpich out on the amber web site, amber.scripps.edu, under the major heading
"Amber-related links" and the minor heading "Running Amber on Linux using
ifc8 and mpich (Provided by Bob Duke)". If you just reference lam stuff
rather than mpich stuff, it should basically work. In the interest of
hopefully making this problem go away longer term, I have constructed a
linux_p4.ifort.lam which you can stick under amber8/src/pmemd/config_data;
then under the pmemd dir do a "./configure linux_p4 ifort lam" and answer
the question about where the lam home is. This should work. I will send
you this file as gzipped text separately for your convenience, but the text
is:

+++
setenv LAM_INCLUDE $LAM_HOME/include
setenv LAM_LIBDIR $LAM_HOME/lib
setenv MPILIB "-L$LAM_LIBDIR -llamf77mpi -lmpi -llam"
setenv PREPROCFLAGS "-DMPI -DSLOW_NONBLOCKING_MPI -DDIRFRC_VECT_OPT"
setenv CPP "/lib/cpp -traditional -I$LAM_INCLUDE"
setenv OPT_LO "ifort -c -auto -tpp7 -mp1 -O0"
setenv OPT_MED "ifort -c -auto -tpp7 -mp1 -O2"
setenv OPT_HI "ifort -c -auto -tpp7 -xW -mp1 -ip -O3"
setenv LOAD "ifort"
setenv LOADLIB " -limf -lsvml $MPILIB" 

----
Regards - Bob Duke

----- Original Message ----- 
From: "Ross Walker" <ross_at_rosswalker.co.uk>
To: <amber_at_scripps.edu>
Sent: Wednesday, February 23, 2005 1:02 PM
Subject: RE: AMBER: RE:Amber 8 with lam compilation problem

>> Thx very much it is working :) . I understand that i must have 2
>> ambers on
>> my computers when i want run seriall jobs and somtimes run
>> parallel , but i
>> must have two different AMBERHOME variables.  How to solve this
>> not setting
>> it always when i want to run seriall ? . Can i add somehow AMBERHOME2
>> variable ?
>
> You don't actually need $AMBERHOME set to run sander. It is only actually
> required for Leap. I just use it in my examples as a kind of generic amber
> path. If you use the full path in your scripts (always a good idea in my
> opinion) then you won't have any problems. Alternatively you could just 
> copy
> the parallel sander and sander.LES executables  out of the parallel
> compilation directory and put them in the serial directory called 
> sander.mpi
> and sander.LES.mpi or something similar. Then just remember to use these
> names instead of just sander when running in parallel.
>
>> Is it possible to compile pmemd with intel fortran 8.1 compiler
>> ?? becouse i
>> cant configure it .
>
> Yes, although you need a special configure script. There is a note posted 
> on
> the archive somewhere about it. Check http://amber.ch.ic.ac.uk/archive.
> There may also be something on the amber website about it but I forget
> where. If you can't find it in the archive or on the website post a 
> message
> to the mailing list again and Bob Duke will probably help you out.
>
> All the best
> Ross
>
> /\
> \/
> |\oss Walker
>
> | Department of Molecular Biology TPC15 |
> | The Scripps Research Institute |
> | Tel:- +1 858 784 8889 | EMail:- ross_at_rosswalker.co.uk |
> | http://www.rosswalker.co.uk/ | PGP Key available on request |
>
> Note: Electronic Mail is not secure, has no guarantee of delivery, may not
> be read every day, and should not be used for urgent or sensitive issues.
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> 
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu