AMBER: Problems with extra points on parralellized linux and sgi machine

From: Sarah Wittkopp <tschampe.sunchem.chem.uga.edu>
Date: Fri, 15 Aug 2003 03:07:11 -0400 (EDT)

Hi,

A while ago I recieved a bugfix, which solved my problem and works
perfectly in an unparallelized environment on a linux pc. The problem
pertained to the error "fill_bonded: max13 exceeded!!".

The system is a sugar, with lone pairs on all oxygens, in a box of tip5p
water (although the same result is obtained when tip3p water is used). I
have tried to run the simulation on both a single cpu as well as
in a parallelized environment. Although the bugfix was added and the code
was recompiled the same way in both cases, I recieve the error only in the
parallelized environment, while the simulation runs fine on the single
cpu:

(the end of the NPT simulation in AMBER7 with ewald)

     Cutoff= 8.000 Tol =0.100E-04
     Ewald Coefficient = 0.34864
     Interpolation order = 4
| PLEVEL = 1: runmd parallelization, no EKCMR
 fill_bonded: max13 exceeded!!

I have tried both the SGI IRIX platform and redhat linux for the
parallelized environment, while the single cpu is under the redhat linux
OS.

I have inlcuded the bugfix below, although it can also be found in the
AMBER archive contributed by Prof. David Case.

Any help would be appreciated,
Thanks
Sarah


*** extra_pts.f 2002/01/11 17:21:17 7.25
--- extra_pts.f 2002/08/15 22:57:53
***************
*** 126,132 ****
  #ifndef LES
        call allocate_frames(numextra,ifrtyp,iatcen,inumep,
       + iepfr,ifrst,imid,ithrd,leploc)
! max14 = 2*(nphih+nphia+ndper)
        call allocate_14nb(inb_14,max14)
  #endif
  c
--- 126,132 ----
  #ifndef LES
        call allocate_frames(numextra,ifrtyp,iatcen,inumep,
       + iepfr,ifrst,imid,ithrd,leploc)
! max14 = 3*(nphih+nphia+ndper)
        call allocate_14nb(inb_14,max14)
  #endif
  c
***************
*** 193,201 ****
        call get_istack(i_numnghbr,3*natom)
        call get_istack(i_epowner,natom)
        max11 = natom+numextra
! max12 = 2*(nbonh+nbona+nbper)
! max13 = 2*(ntheth+ntheta+ngper)
! max14 = 2*(nphih+nphia+ndper)
        maxa = max(max11,max12,max13,max14)
        call get_istack(i_11,2*max11)
        call get_istack(i_12,2*max12)
--- 193,201 ----
        call get_istack(i_numnghbr,3*natom)
        call get_istack(i_epowner,natom)
        max11 = natom+numextra
! max12 = 3*(nbonh+nbona+nbper)
! max13 = 3*(ntheth+ntheta+ngper)
! max14 = 3*(nphih+nphia+ndper)
        maxa = max(max11,max12,max13,max14)
        call get_istack(i_11,2*max11)
        call get_istack(i_12,2*max12)

***********************************************


-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Fri Aug 15 2003 - 08:53:00 PDT
Custom Search