 | |||||||||||||||
 |
 |
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
 |
AMBER Archive (2008)
Subject: Re: AMBER: Non bond list error
From: Robert Duke (rduke_at_email.unc.edu)
Date: Fri Oct 24 2008 - 13:47:33 CDT
- Next message: Robert Duke: "Re: AMBER: Non bond list error"
- Previous message: Francesco Pietra: "AMBER: Atom names CL and four characters"
- In reply to: neville forlemu: "Re: AMBER: Non bond list error"
- Next in thread: Robert Duke: "Re: AMBER: Non bond list error"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
This message has been processed by Symantec's AntiVirus Technology.
Unknown00000000.data was not scanned for viruses because too many nested levels of files were found.
For more information on antivirus tips and technology, visit
http://ses.symantec.com/
attached mail follows:
Okay, WHOA. You are getting a SIGBUS error here. That is caused by either a memory addressing error or reference to an incorrectly aligned address for a datatype - depends on architecture issues for the 2nd. This indicates, most likely, an incorrectly built executable (linked to the wrong stuff perhaps), or perhaps a compiler problem (less likely). Then, looking at your new mdin, you are invoking a generalized Born implicit solvent run (igb = 1), whereas before you were attempting to use pme with explicit solvent under periodic boundary conditions. SO which did you really build - a solvated system for use with pme, or an unsolvated system for use with generalized Born? I would recommend:
1) Test sander.MPI and pmemd by running the test suites that come with amber (read the manual and README's as to how to do this stuff...)
2) Go to the ambermd.org and follow the tutorials link to get to some excellent tutorials written by Ross and others. Play with these for several days, and confirm that they do what they ought to be doing.
3) Then look carefully at how you built your own system and try again.
Regards - Bob Duke
----- Original Message -----
From: neville forlemu
To: amber_at_scripps.edu
Sent: Friday, October 24, 2008 2:12 PM
Subject: Re: AMBER: Non bond list error
Hi
I thought we had coz the executable is in one of the amber directories (/opt/brinsoft/amber8/exe/pmemd) on my machine.
So that is what I get
mpirun -np 4 $AMBERHOME/exe/pmemd -O -i minwat.in -o minwat.out -p rldhn_sol.top -c rldhn_sol.crd -r minwat.resrst -ref rldhn_sol.crd
MPI: On host brinsvr1, Program /opt/brinsoft/amber8/exe/pmemd, Rank 0, Process 394311 received signal SIGBUS(10)
MPI: --------stack traceback-------
PC: 0x4045080 MPI_SGI_stacktraceback in /usr/lib32/libmpi.so
PC: 0x40455a8 first_arriver_handler in /usr/lib32/libmpi.so
PC: 0x4045858 slave_sig_handler in /usr/lib32/libmpi.so
PC: 0xfaf064c _sigtramp in /usr/lib32/libc.so.1
PC: 0x1001d6a8 SETUP_CRD_IDX_LST_TBL.in.EW_DIRECT_CIT_MODULE in /opt/brinsoft/amber8/exe/pmemd
PC: 0x10079530 DO_INITIAL_ATOM_DIVISION_CIT.in.ALLTASKS_SETUP_MODULE in /opt/brinsoft/amber8/exe/pmemd
PC: 0x10078218 ALLTASKS_SETUP.in.ALLTASKS_SETUP_MODULE in /opt/brinsoft/amber8/exe/pmemd
PC: 0x1005a1f4 pmemd in /opt/brinsoft/amber8/exe/pmemd
PC: 0xad69ef4 main in /usr/lib32/libftn.so
sh: dbx: not found
MPI: -----stack traceback ends-----
MPI: On host brinsvr1, Program /opt/brinsoft/amber8/exe/pmemd, Rank 0, Process 394311: Dumping core on signal SIGBUS(10) into directory /usr/people/forlemu/chanel/ambtrial/amb3
MPI: Program /opt/brinsoft/amber8/exe/pmemd, Rank 2, Process 394464: Core dump on signal SIGBUS(10) suppressed.
MPI: Program /opt/brinsoft/amber8/exe/pmemd, Rank 3, Process 394504: Core dump on signal SIGBUS(10) suppressed.
MPI: MPI_COMM_WORLD rank 1 has terminated without calling MPI_Finalize()
MPI: aborting job
MPI: Received signal 9
However I also changed my input file to this
# test of minimization
&cntrl
maxcyc=400, imin=1, cut=12.0, igb=1, ntb=0, ntpr=10,
/
And used sanders again, and the minimization seems to be working ok.
But I have to now worry about fixing some residues and letting others move, with the above input.
I appreciate your help
Neville
--- On Fri, 10/24/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 11:40 AM
Have you built pmemd? It is a separate step after the general build of amber. If you have built it, is it somewhere that it can be found? Here you just reference it as "pmemd" with no path. I generally use something like $AMBERHOME/exe/pmemd, ASSUMING it has been built...
Regards - Bob Duke
----- Original Message -----
From: neville forlemu
To: amber_at_scripps.edu
Sent: Friday, October 24, 2008 12:34 PM
Subject: RE: AMBER: Non bond list error
Hi Ross,
I did change the parameters you suggested to their defaults, and also changed the forcefield type, but got the same error from sanders.
However when I try pmemd this is the error I got
mpirun -np 4 pmemd -O -i minwat.in -o minwat.out -p rldhn_sol.top -crldhn_sol.crd -r minwat.resrst -ref rldhn_sol.crd
MPI: brinsvr1: 0x6c6bffff00005964: /usr/local/gnu/bin/bash: line 1: exec: pmemd: not found
MPI: could not run executable (all_signal.c:206)
Neville
--- On Fri, 10/24/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 1:12 AM
Hi Neville
You have a valid reason for this:
scnb = 1.0, scee = 1.0, ??????
You should NEVER EVER mess with the defaults of these values unless you know exactly what you are doing, i.e. using a non standard force field that uses different 1-4 scaling factors. Setting these to 1.0 for the AMBER FF series force fields will invalidate ALL of your results.
Note you also probably do not want to be using FF99. I suggest using FF99SB or FF03.
You should also not be messing with nsnb - leave that at the default as well.
As for the Non bond list overflow problem try setting cut=8.0 and see if it goes away.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 10:13 PM
To: amber_at_scripps.edu
Subject: RE: AMBER: Non bond list error
Hi Ross
I am using amber8 and trying to use sanders to minimize waters around this huge protein of about 20000 atoms, then perform a short md on some residues in the protein.
Constant Volume Minimization
# Control section
&cntrl
ntwx = 500, ntpr = 500, ntwr = 500,
scnb = 1.0, scee = 1.0, nsnb = 25, dielc = 1, cut = 12.0,
ntb = 1, ntc = 2, ntf = 2,
maxcyc = 10000, ntmin = 1, ncyc = 1000, drms = 0.0001,
ntp = 0,
ibelly = 0, ntr = 1,
imin = 1,
&end
Group Input for restrained atoms
5.0
RES 1 1332
END
END
Thanks
Neville
Also below is the xleap file preparation
Welcome to LEaP!
(no leaprc in search path)
> source leaprc.ff99
----- Source: /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99
----- Source of /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99 done
Log file: ./leap.log
Loading parameters: /opt/brinsoft/amber8/dat/leap/parm/parm99.dat
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_nucleic94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_amino94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminoct94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminont94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/ions94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/solvents.lib
> x = loadpdb rldhN.pdb
Loading PDB file: ./rldhN.pdb
(starting new molecule for chain B)
(starting new molecule for chain C)
(starting new molecule for chain D)
total atoms in file: 10236
Leap added 10556 missing atoms according to residue templates:
10556 H / lone pairs
> savepdb x rld.pdb
Writing pdb file: rld.pdb
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
> y = loadpdb rld.pdb
Loading PDB file: ./rld.pdb
total atoms in file: 20792
> addions y Cl- 0
8 Cl- ions required to neutralize.
Adding 8 counter ions to "y" using 1A grid
Grid extends from solute vdw + 2.47 to 8.47
Resolution: 1.00 Angstrom.
grid build: 1 sec
(no solvent present)
Calculating grid charges
charges: 251 sec
Placed Cl- in y at (5.12, -4.35, -51.18).
Placed Cl- in y at (4.12, 4.65, 0.82).
Placed Cl- in y at (13.12, -39.35, -35.18).
Placed Cl- in y at (-3.88, 16.65, -2.18).
Placed Cl- in y at (29.12, -0.35, -20.18).
Placed Cl- in y at (-8.88, -29.35, 3.82).
Placed Cl- in y at (-9.88, 32.65, -53.18).
Placed Cl- in y at (-27.88, -7.35, -31.18).
Done adding ions.
> solvateBox y TIP3PBOX 10
Solute vdw bounding box: 76.168 91.281 70.396
Total bounding box for atom centers: 96.168 111.281 90.396
Solvent unit box: 18.774 18.774 18.774
Total vdw box size: 99.135 92.364 120.398 angstroms.
Volume: 1102429.931 A^3
Total mass 702324.416 amu, Density 1.058 g/cc
> saveamberparm y rldhn_sol.top rldhn_sol.crd
Checking Unit.
Building topology.
Building atom parameters.
Building bond parameters.
Building angle parameters.
Building proper torsion parameters.
Building improper torsion parameters.
total 3824 improper torsions applied
Building H-Bond parameters.
Not Marking per-residue atom chain types.
Marking per-residue atom chain types.
(Residues lacking connect0/connect1 -
these don't have chain types marked:
--- On Thu, 10/23/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 11:56 PM
Hi Neville,
Can you post your input file and some details about your system (number of atoms, type of simulation, gas phase, periodic boundaries etc) as well as the version of AMBER you are using.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 9:48 PM
To: amber_at_scripps.edu; rduke_at_email.unc.edu
Subject: Re: AMBER: Non bond list error
Hello,
Could some one explain to me what this error means
* NB pairs 7104 103754005 exceeds capacity ( 103754298) 0
SIZE OF NONBOND LIST = 103754298
SANDER BOMB in subroutine nonbond_list
Non bond list overflow!
check MAXPR in locmem.f
I am trying to run sander for energy minimization, but keep running into this problem.
Thanks
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show that you didn't... (oh, oops). So it is memory corruption. What I don't understand is why you are not dying with some sort of "out of memory" error from sander, associated with asking for more memory than is available. For pmemd, anywhere I allocate dynamic memory, I check for a success return code, so the way you should experience running out of memory there is to get an explicit error message. Because sander has a preallocated memory pool strategy, I suspect that other things are possible... Bottom line on
all this - I think it is a good idea to not run more than roughly 100,000 atoms on a single processor, especially for sander. And if you run it on 4 processors but they all share the same limited physical memory, you may also hit trouble. I attached a graphic on pmemd memory requirements - a jpg so it should be widely viewable. My rule of thumb for pmemd is that 4 processors, each with 1 GB of actual physical memory, can handle up to 1 million atoms with the default 8 angstrom cutoff. Sander will take more. There are also buffer space considerations in an mpi application (within mpi itself, not in the app), that
further muddy the waters, but following this guideline you should be safe. Regards - Bob ----- Original Message ----- From: "Robert Duke" <rduke_at_email.unc.edu> To: <amber_at_scripps.edu> Sent: Thursday, October 23, 2008 4:14 PM Subject: Re: AMBER: Non bond list error > As Ross will tell you too: > 1) Don't increase cut to 12, leave it at the default (of 8) > 2) Run this on at least 4 processors using the MPI version of pmemd or > sander (I
know you are using sander here; pmemd requires less memory). > Even higher processor counts will reduce your risk of memory overflow > further. Your pairlist went negative because you incremented it past a 31 > bit digit; with the commonly used integer format on computers these days > (twos-complement), this results in a negative number (and is clearly an > error condition). Is this memory usage reasonable for the size problem > you have? Well, that cutoff plus skin will produce about 552 pairs per > atom. If you had 1,000,000 atoms (and you are close), that would be > 552,000,000 pairs. Not enough to overflow the list
counter. BUT that is > 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list > alone. Most machines, you are pushing it to get much over 1.5 GB for the > application (I have not looked recently, so that is off the top of my > head). With true 32 bit executables, you are out of address space; with > the newer 64 bit chips, you have bits to specify more than 2 GB of > addresses, but you may not have enough actual memory. And remember that > the pairlist is only part of your memory consumption. No resource is > infinite on a computer... > Regards - Bob
Duke > ----- Original Message ----- > From: "Wang,Ying" <wangying_at_ufl.edu> > To: <amber_at_scripps.edu> > Sent: Thursday, October 23, 2008 3:26 PM > Subject: RE: AMBER: Non bond list error > > >> Hi, Ross, >> >> Thanks a lot! >> >> My input file is as below: >> 50ps MD with res >> &cntrl >> imin = 0, >>
irest = 0, >> ntx = 1, >> ntb = 1, >> cut = 12, >> ntr = 1, >> ntc = 2, >> ntf = 2, >> tempi = 0.0, >> temp0 = 300.0, >> ntt = 3, >> gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001 >> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> nmropt=1 >> / >> &wt TYPE='TEMP0', istep1=0, istep2=50000, >> value1=0.1, value2=300.0, / >> &wt TYPE='END' / >> Keep system fixed with weak restraints >> 20.0 >> RES 1 5076 >> END >> END >> >> and the NPT is as below: >> >> NPT: 50ps MD >> &cntrl >> imin = 0, irest = 1, ntx = 7,
>> ntb = 2, pres0 = 1.0, ntp = 1, >> taup = 2.0, >> cut = 12, ntr = 1, >> ntc = 2, ntf = 2, >> tempi = 300.0, temp0 = 300.0, >> ntt = 3, gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001, >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >> / >> Keep fixed with weak restraints >> 20.0 >> RES 217 954 >> END >> Keep fixed with weak restraints
>> 20.0 >> RES 1909 2646 >> END >> Keep fixed with weak restraints >> 20.0 >> RES 3601 4338 >> END >> res also >> 5.0 >> RES 955 1692 >> END >> res also >> 5.0 >> RES 2647 3384 >> END >> res also >> 5.0 >> RES 4339 5076 >>
END >> END >> >> >> >> Thanks again!!!!!!!!!!!!!!! >> >> >> >> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker <ross_at_rosswalker.co.uk> >> wrote: >> >>> Hi Wang, >>> >>> 800K atoms is pretty large and while sander / pmemd should support this >>> size
>>> (I think 999,999 is the limit right now due to file formatting) you may >>> run >>> into problems that haven't been seen before. >>> >>> It's not obvious what is going wrong in your case but the numbers don't >>> make >>> any sense (a negative capacity!) which suggests either memory corruption >>> through an array overflow or the number of pairs is larger than a signed >>> integer and is overflowing. Even at 800K atoms you shouldn't
have this >>> many >>> pairs though. Can you post your input file so we can take a look? I >>> suspect >>> you have cut set too high or perhaps are not running PME etc. >>> >>> All the best >>> Ross >>> >>>> -----Original Message----- >>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On >>>> Behalf >>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM >>>> To: amber_at_scripps.edu >>>> Subject: AMBER: Non bond list error >>>> >>>> Hi, Dear AMBERs, >>>> >>>> I meet a problem when I run a simulation of a system consist of >>>> 799889 atoms. >>>> >>>> * NB pairs 451 0 exceeds capacity ( >>>>
-28510921) 7 >>>> SIZE OF NONBOND LIST = -28510921 >>>> SANDER BOMB in subroutine nonbond_list >>>> Non bond list overflow! >>>> check MAXPR in locmem.f >>>> >>>> Could anyone tell me what's happen? >>>> >>>> Thanks a lot! >>>> >>>> ----------------------------------------------------------------------- >>>> The AMBER Mail Reflector
>>>> To post, send mail to amber_at_scripps.edu >>>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>>> to majordomo_at_scripps.edu >>> >>> ----------------------------------------------------------------------- >>> The AMBER Mail Reflector >>> To post, send mail to amber_at_scripps.edu >>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>> to
majordomo_at_scripps.edu >>> >>> >> >> >> >> -- >> Wang,Ying >> >> ----------------------------------------------------------------------- >> The AMBER Mail Reflector >> To post, send mail to amber_at_scripps.edu >> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >> to majordomo_at_scripps.edu
>> > > ----------------------------------------------------------------------- > The AMBER Mail Reflector > To post, send mail to amber_at_scripps.edu > To unsubscribe, send "unsubscribe amber" (in the *body* of the email) > to majordomo_at_scripps.edu >
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show that you didn't... (oh, oops). So it is memory corruption. What I don't understand is why you are not dying with some sort of "out of memory" error from sander, associated with asking for more memory than is available. For pmemd, anywhere I allocate dynamic memory, I check for a success return code, so the way you should experience running out of memory there is to get an explicit error message. Because sander has a preallocated memory pool strategy, I suspect that other things are possible... Bottom line
on all this - I think it is a good idea to not run more than roughly 100,000 atoms on a single processor, especially for sander. And if you run it on 4 processors but they all share the same limited physical memory, you may also hit trouble. I attached a graphic on pmemd memory requirements - a jpg so it should be widely viewable. My rule of thumb for pmemd is that 4 processors, each with 1 GB of actual physical memory, can handle up to 1 million atoms with the default 8 angstrom cutoff. Sander will take more. There are also buffer space considerations in an mpi application (within mpi itself, not in the app), that
further muddy the waters, but following this guideline you should be safe. Regards - Bob ----- Original Message ----- From: "Robert Duke" <rduke_at_email.unc.edu> To: <amber_at_scripps.edu> Sent: Thursday, October 23, 2008 4:14 PM Subject: Re: AMBER: Non bond list error > As Ross will tell you too: > 1) Don't increase cut to 12, leave it at the default (of 8) > 2) Run this on at least 4 processors using the MPI version of pmemd or > sander (I
know you are using sander here; pmemd requires less memory). > Even higher processor counts will reduce your risk of memory overflow > further. Your pairlist went negative because you incremented it past a 31 > bit digit; with the commonly used integer format on computers these days > (twos-complement), this results in a negative number (and is clearly an > error condition). Is this memory usage reasonable for the size problem > you have? Well, that cutoff plus skin will produce about 552 pairs per > atom. If you had 1,000,000 atoms (and you are close), that would be > 552,000,000 pairs. Not enough to overflow the list
counter. BUT that is > 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list > alone. Most machines, you are pushing it to get much over 1.5 GB for the > application (I have not looked recently, so that is off the top of my > head). With true 32 bit executables, you are out of address space; with > the newer 64 bit chips, you have bits to specify more than 2 GB of > addresses, but you may not have enough actual memory. And remember that > the pairlist is only part of your memory consumption. No resource is > infinite on a computer... > Regards - Bob Duke
> ----- Original Message ----- > From: "Wang,Ying" <wangying_at_ufl.edu> > To: <amber_at_scripps.edu> > Sent: Thursday, October 23, 2008 3:26 PM > Subject: RE: AMBER: Non bond list error > > >> Hi, Ross, >> >> Thanks a lot! >> >> My input file is as below: >> 50ps MD with res >> &cntrl >> imin = 0, >>
irest = 0, >> ntx = 1, >> ntb = 1, >> cut = 12, >> ntr = 1, >> ntc = 2, >> ntf = 2, >> tempi = 0.0, >> temp0 = 300.0, >> ntt = 3, >> gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001 >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >>
nmropt=1 >> / >> &wt TYPE='TEMP0', istep1=0, istep2=50000, >> value1=0.1, value2=300.0, / >> &wt TYPE='END' / >> Keep system fixed with weak restraints >> 20.0 >> RES 1 5076 >> END >> END >> >> and the NPT is as below: >> >> NPT: 50ps MD >> &cntrl >> imin = 0, irest = 1, ntx = 7, >>
ntb = 2, pres0 = 1.0, ntp = 1, >> taup = 2.0, >> cut = 12, ntr = 1, >> ntc = 2, ntf = 2, >> tempi = 300.0, temp0 = 300.0, >> ntt = 3, gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001, >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >> / >> Keep fixed with weak restraints >> 20.0 >> RES 217 954 >> END >> Keep fixed with weak restraints >> 20.0
>> RES 1909 2646 >> END >> Keep fixed with weak restraints >> 20.0 >> RES 3601 4338 >> END >> res also >> 5.0 >> RES 955 1692 >> END >> res also >> 5.0 >> RES 2647 3384 >> END >> res also >> 5.0 >> RES 4339 5076 >> END
>> END >> >> >> >> Thanks again!!!!!!!!!!!!!!! >> >> >> >> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker <ross_at_rosswalker.co.uk> >> wrote: >> >>> Hi Wang, >>> >>> 800K atoms is pretty large and while sander / pmemd should support this >>> size >>> (I think
999,999 is the limit right now due to file formatting) you may >>> run >>> into problems that haven't been seen before. >>> >>> It's not obvious what is going wrong in your case but the numbers don't >>> make >>> any sense (a negative capacity!) which suggests either memory corruption >>> through an array overflow or the number of pairs is larger than a signed >>> integer and is overflowing. Even at 800K atoms you shouldn't have this
>>> many >>> pairs though. Can you post your input file so we can take a look? I >>> suspect >>> you have cut set too high or perhaps are not running PME etc. >>> >>> All the best >>> Ross >>> >>>> -----Original Message----- >>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On >>>> Behalf >>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM >>>> To: amber_at_scripps.edu >>>> Subject: AMBER: Non bond list error >>>> >>>> Hi, Dear AMBERs, >>>> >>>> I meet a problem when I run a simulation of a system consist of >>>> 799889 atoms. >>>> >>>> * NB pairs 451 0 exceeds capacity ( >>>> -28510921) 7
>>>> SIZE OF NONBOND LIST = -28510921 >>>> SANDER BOMB in subroutine nonbond_list >>>> Non bond list overflow! >>>> check MAXPR in locmem.f >>>> >>>> Could anyone tell me what's happen? >>>> >>>> Thanks a lot! >>>> >>>> ----------------------------------------------------------------------- >>>> The AMBER Mail Reflector >>>> To post,
send mail to amber_at_scripps.edu >>>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>>> to majordomo_at_scripps.edu >>> >>> ----------------------------------------------------------------------- >>> The AMBER Mail Reflector >>> To post, send mail to amber_at_scripps.edu >>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>> to majordomo_at_scripps.edu
>>> >>> >> >> >> >> -- >> Wang,Ying >> >> ----------------------------------------------------------------------- >> The AMBER Mail Reflector >> To post, send mail to amber_at_scripps.edu >> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >> to majordomo_at_scripps.edu >>
> > ----------------------------------------------------------------------- > The AMBER Mail Reflector > To post, send mail to amber_at_scripps.edu > To unsubscribe, send "unsubscribe amber" (in the *body* of the email) > to majordomo_at_scripps.edu >
--- On Fri, 10/24/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 11:40 AM
Have you built pmemd? It is a separate step after the general build of amber. If you have built it, is it somewhere that it can be found? Here you just reference it as "pmemd" with no path. I generally use something like $AMBERHOME/exe/pmemd, ASSUMING it has been built...
Regards - Bob Duke
----- Original Message -----
From: neville forlemu
To: amber_at_scripps.edu
Sent: Friday, October 24, 2008 12:34 PM
Subject: RE: AMBER: Non bond list error
Hi Ross,
I did change the parameters you suggested to their defaults, and also changed the forcefield type, but got the same error from sanders.
However when I try pmemd this is the error I got
mpirun -np 4 pmemd -O -i minwat.in -o minwat.out -p rldhn_sol.top -crldhn_sol.crd -r minwat.resrst -ref rldhn_sol.crd
MPI: brinsvr1: 0x6c6bffff00005964: /usr/local/gnu/bin/bash: line 1: exec: pmemd: not found
MPI: could not run executable (all_signal.c:206)
Neville
--- On Fri, 10/24/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 1:12 AM
Hi Neville
You have a valid reason for this:
scnb = 1.0, scee = 1.0, ??????
You should NEVER EVER mess with the defaults of these values unless you know exactly what you are doing, i.e. using a non standard force field that uses different 1-4 scaling factors. Setting these to 1.0 for the AMBER FF series force fields will invalidate ALL of your results.
Note you also probably do not want to be using FF99. I suggest using FF99SB or FF03.
You should also not be messing with nsnb - leave that at the default as well.
As for the Non bond list overflow problem try setting cut=8.0 and see if it goes away.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 10:13 PM
To: amber_at_scripps.edu
Subject: RE: AMBER: Non bond list error
Hi Ross
I am using amber8 and trying to use sanders to minimize waters around this huge protein of about 20000 atoms, then perform a short md on some residues in the protein.
Constant Volume Minimization
# Control section
&cntrl
ntwx = 500, ntpr = 500, ntwr = 500,
scnb = 1.0, scee = 1.0, nsnb = 25, dielc = 1, cut = 12.0,
ntb = 1, ntc = 2, ntf = 2,
maxcyc = 10000, ntmin = 1, ncyc = 1000, drms = 0.0001,
ntp = 0,
ibelly = 0, ntr = 1,
imin = 1,
&end
Group Input for restrained atoms
5.0
RES 1 1332
END
END
Thanks
Neville
Also below is the xleap file preparation
Welcome to LEaP!
(no leaprc in search path)
> source leaprc.ff99
----- Source: /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99
----- Source of /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99 done
Log file: ./leap.log
Loading parameters: /opt/brinsoft/amber8/dat/leap/parm/parm99.dat
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_nucleic94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_amino94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminoct94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminont94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/ions94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/solvents.lib
> x = loadpdb rldhN.pdb
Loading PDB file: ./rldhN.pdb
(starting new molecule for chain B)
(starting new molecule for chain C)
(starting new molecule for chain D)
total atoms in file: 10236
Leap added 10556 missing atoms according to residue templates:
10556 H / lone pairs
> savepdb x rld.pdb
Writing pdb file: rld.pdb
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
> y = loadpdb rld.pdb
Loading PDB file: ./rld.pdb
total atoms in file: 20792
> addions y Cl- 0
8 Cl- ions required to neutralize.
Adding 8 counter ions to "y" using 1A grid
Grid extends from solute vdw + 2.47 to 8.47
Resolution: 1.00 Angstrom.
grid build: 1 sec
(no solvent present)
Calculating grid charges
charges: 251 sec
Placed Cl- in y at (5.12, -4.35, -51.18).
Placed Cl- in y at (4.12, 4.65, 0.82).
Placed Cl- in y at (13.12, -39.35, -35.18).
Placed Cl- in y at (-3.88, 16.65, -2.18).
Placed Cl- in y at (29.12, -0.35, -20.18).
Placed Cl- in y at (-8.88, -29.35, 3.82).
Placed Cl- in y at (-9.88, 32.65, -53.18).
Placed Cl- in y at (-27.88, -7.35, -31.18).
Done adding ions.
> solvateBox y TIP3PBOX 10
Solute vdw bounding box: 76.168 91.281 70.396
Total bounding box for atom centers: 96.168 111.281 90.396
Solvent unit box: 18.774 18.774 18.774
Total vdw box size: 99.135 92.364 120.398 angstroms.
Volume: 1102429.931 A^3
Total mass 702324.416 amu, Density 1.058 g/cc
> saveamberparm y rldhn_sol.top rldhn_sol.crd
Checking Unit.
Building topology.
Building atom parameters.
Building bond parameters.
Building angle parameters.
Building proper torsion parameters.
Building improper torsion parameters.
total 3824 improper torsions applied
Building H-Bond parameters.
Not Marking per-residue atom chain types.
Marking per-residue atom chain types.
(Residues lacking connect0/connect1 -
these don't have chain types marked:
--- On Thu, 10/23/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 11:56 PM
Hi Neville,
Can you post your input file and some details about your system (number of atoms, type of simulation, gas phase, periodic boundaries etc) as well as the version of AMBER you are using.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 9:48 PM
To: amber_at_scripps.edu; rduke_at_email.unc.edu
Subject: Re: AMBER: Non bond list error
Hello,
Could some one explain to me what this error means
* NB pairs 7104 103754005 exceeds capacity ( 103754298) 0
SIZE OF NONBOND LIST = 103754298
SANDER BOMB in subroutine nonbond_list
Non bond list overflow!
check MAXPR in locmem.f
I am trying to run sander for energy minimization, but keep running into this problem.
Thanks
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show that you didn't... (oh, oops). So it is memory corruption. What I don't understand is why you are not dying with some sort of "out of memory" error from sander, associated with asking for more memory than is available. For pmemd, anywhere I allocate dynamic memory, I check for a success return code, so the way you should experience running out of memory there is to get an explicit error message. Because sander has a preallocated memory pool strategy, I suspect that other things are possible... Bottom line on
all this - I think it is a good idea to not run more than roughly 100,000 atoms on a single processor, especially for sander. And if you run it on 4 processors but they all share the same limited physical memory, you may also hit trouble. I attached a graphic on pmemd memory requirements - a jpg so it should be widely viewable. My rule of thumb for pmemd is that 4 processors, each with 1 GB of actual physical memory, can handle up to 1 million atoms with the default 8 angstrom cutoff. Sander will take more. There are also buffer space considerations in an mpi application (within mpi itself, not in the app), that
further muddy the waters, but following this guideline you should be safe. Regards - Bob ----- Original Message ----- From: "Robert Duke" <rduke_at_email.unc.edu> To: <amber_at_scripps.edu> Sent: Thursday, October 23, 2008 4:14 PM Subject: Re: AMBER: Non bond list error > As Ross will tell you too: > 1) Don't increase cut to 12, leave it at the default (of 8) > 2) Run this on at least 4 processors using the MPI version of pmemd or > sander (I
know you are using sander here; pmemd requires less memory). > Even higher processor counts will reduce your risk of memory overflow > further. Your pairlist went negative because you incremented it past a 31 > bit digit; with the commonly used integer format on computers these days > (twos-complement), this results in a negative number (and is clearly an > error condition). Is this memory usage reasonable for the size problem > you have? Well, that cutoff plus skin will produce about 552 pairs per > atom. If you had 1,000,000 atoms (and you are close), that would be > 552,000,000 pairs. Not enough to overflow the list
counter. BUT that is > 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list > alone. Most machines, you are pushing it to get much over 1.5 GB for the > application (I have not looked recently, so that is off the top of my > head). With true 32 bit executables, you are out of address space; with > the newer 64 bit chips, you have bits to specify more than 2 GB of > addresses, but you may not have enough actual memory. And remember that > the pairlist is only part of your memory consumption. No resource is > infinite on a computer... > Regards - Bob
Duke > ----- Original Message ----- > From: "Wang,Ying" <wangying_at_ufl.edu> > To: <amber_at_scripps.edu> > Sent: Thursday, October 23, 2008 3:26 PM > Subject: RE: AMBER: Non bond list error > > >> Hi, Ross, >> >> Thanks a lot! >> >> My input file is as below: >> 50ps MD with res >> &cntrl >> imin = 0, >>
irest = 0, >> ntx = 1, >> ntb = 1, >> cut = 12, >> ntr = 1, >> ntc = 2, >> ntf = 2, >> tempi = 0.0, >> temp0 = 300.0, >> ntt = 3, >> gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001 >> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> nmropt=1 >> / >> &wt TYPE='TEMP0', istep1=0, istep2=50000, >> value1=0.1, value2=300.0, / >> &wt TYPE='END' / >> Keep system fixed with weak restraints >> 20.0 >> RES 1 5076 >> END >> END >> >> and the NPT is as below: >> >> NPT: 50ps MD >> &cntrl >> imin = 0, irest = 1, ntx = 7,
>> ntb = 2, pres0 = 1.0, ntp = 1, >> taup = 2.0, >> cut = 12, ntr = 1, >> ntc = 2, ntf = 2, >> tempi = 300.0, temp0 = 300.0, >> ntt = 3, gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001, >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >> / >> Keep fixed with weak restraints >> 20.0 >> RES 217 954 >> END >> Keep fixed with weak restraints
>> 20.0 >> RES 1909 2646 >> END >> Keep fixed with weak restraints >> 20.0 >> RES 3601 4338 >> END >> res also >> 5.0 >> RES 955 1692 >> END >> res also >> 5.0 >> RES 2647 3384 >> END >> res also >> 5.0 >> RES 4339 5076 >>
END >> END >> >> >> >> Thanks again!!!!!!!!!!!!!!! >> >> >> >> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker <ross_at_rosswalker.co.uk> >> wrote: >> >>> Hi Wang, >>> >>> 800K atoms is pretty large and while sander / pmemd should support this >>> size
>>> (I think 999,999 is the limit right now due to file formatting) you may >>> run >>> into problems that haven't been seen before. >>> >>> It's not obvious what is going wrong in your case but the numbers don't >>> make >>> any sense (a negative capacity!) which suggests either memory corruption >>> through an array overflow or the number of pairs is larger than a signed >>> integer and is overflowing. Even at 800K atoms you shouldn't
have this >>> many >>> pairs though. Can you post your input file so we can take a look? I >>> suspect >>> you have cut set too high or perhaps are not running PME etc. >>> >>> All the best >>> Ross >>> >>>> -----Original Message----- >>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On >>>> Behalf >>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM >>>> To: amber_at_scripps.edu >>>> Subject: AMBER: Non bond list error >>>> >>>> Hi, Dear AMBERs, >>>> >>>> I meet a problem when I run a simulation of a system consist of >>>> 799889 atoms. >>>> >>>> * NB pairs 451 0 exceeds capacity ( >>>>
-28510921) 7 >>>> SIZE OF NONBOND LIST = -28510921 >>>> SANDER BOMB in subroutine nonbond_list >>>> Non bond list overflow! >>>> check MAXPR in locmem.f >>>> >>>> Could anyone tell me what's happen? >>>> >>>> Thanks a lot! >>>> >>>> ----------------------------------------------------------------------- >>>> The AMBER Mail Reflector
>>>> To post, send mail to amber_at_scripps.edu >>>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>>> to majordomo_at_scripps.edu >>> >>> ----------------------------------------------------------------------- >>> The AMBER Mail Reflector >>> To post, send mail to amber_at_scripps.edu >>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>> to
majordomo_at_scripps.edu >>> >>> >> >> >> >> -- >> Wang,Ying >> >> ----------------------------------------------------------------------- >> The AMBER Mail Reflector >> To post, send mail to amber_at_scripps.edu >> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >> to majordomo_at_scripps.edu
>> > > ----------------------------------------------------------------------- > The AMBER Mail Reflector > To post, send mail to amber_at_scripps.edu > To unsubscribe, send "unsubscribe amber" (in the *body* of the email) > to majordomo_at_scripps.edu >
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show that you didn't... (oh, oops). So it is memory corruption. What I don't understand is why you are not dying with some sort of "out of memory" error from sander, associated with asking for more memory than is available. For pmemd, anywhere I allocate dynamic memory, I check for a success return code, so the way you should experience running out of memory there is to get an explicit error message. Because sander has a preallocated memory pool strategy, I suspect that other things are possible... Bottom line
on all this - I think it is a good idea to not run more than roughly 100,000 atoms on a single processor, especially for sander. And if you run it on 4 processors but they all share the same limited physical memory, you may also hit trouble. I attached a graphic on pmemd memory requirements - a jpg so it should be widely viewable. My rule of thumb for pmemd is that 4 processors, each with 1 GB of actual physical memory, can handle up to 1 million atoms with the default 8 angstrom cutoff. Sander will take more. There are also buffer space considerations in an mpi application (within mpi itself, not in the app), that
further muddy the waters, but following this guideline you should be safe. Regards - Bob ----- Original Message ----- From: "Robert Duke" <rduke_at_email.unc.edu> To: <amber_at_scripps.edu> Sent: Thursday, October 23, 2008 4:14 PM Subject: Re: AMBER: Non bond list error > As Ross will tell you too: > 1) Don't increase cut to 12, leave it at the default (of 8) > 2) Run this on at least 4 processors using the MPI version of pmemd or > sander (I
know you are using sander here; pmemd requires less memory). > Even higher processor counts will reduce your risk of memory overflow > further. Your pairlist went negative because you incremented it past a 31 > bit digit; with the commonly used integer format on computers these days > (twos-complement), this results in a negative number (and is clearly an > error condition). Is this memory usage reasonable for the size problem > you have? Well, that cutoff plus skin will produce about 552 pairs per > atom. If you had 1,000,000 atoms (and you are close), that would be > 552,000,000 pairs. Not enough to overflow the list
counter. BUT that is > 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list > alone. Most machines, you are pushing it to get much over 1.5 GB for the > application (I have not looked recently, so that is off the top of my > head). With true 32 bit executables, you are out of address space; with > the newer 64 bit chips, you have bits to specify more than 2 GB of > addresses, but you may not have enough actual memory. And remember that > the pairlist is only part of your memory consumption. No resource is > infinite on a computer... > Regards - Bob Duke
> ----- Original Message ----- > From: "Wang,Ying" <wangying_at_ufl.edu> > To: <amber_at_scripps.edu> > Sent: Thursday, October 23, 2008 3:26 PM > Subject: RE: AMBER: Non bond list error > > >> Hi, Ross, >> >> Thanks a lot! >> >> My input file is as below: >> 50ps MD with res >> &cntrl >> imin = 0, >>
irest = 0, >> ntx = 1, >> ntb = 1, >> cut = 12, >> ntr = 1, >> ntc = 2, >> ntf = 2, >> tempi = 0.0, >> temp0 = 300.0, >> ntt = 3, >> gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001 >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >>
nmropt=1 >> / >> &wt TYPE='TEMP0', istep1=0, istep2=50000, >> value1=0.1, value2=300.0, / >> &wt TYPE='END' / >> Keep system fixed with weak restraints >> 20.0 >> RES 1 5076 >> END >> END >> >> and the NPT is as below: >> >> NPT: 50ps MD >> &cntrl >> imin = 0, irest = 1, ntx = 7, >>
ntb = 2, pres0 = 1.0, ntp = 1, >> taup = 2.0, >> cut = 12, ntr = 1, >> ntc = 2, ntf = 2, >> tempi = 300.0, temp0 = 300.0, >> ntt = 3, gamma_ln = 2.0, >> nstlim = 50000, dt = 0.001, >> ntpr = 1000, ntwx = 1000, ntwr = 1000 >> / >> Keep fixed with weak restraints >> 20.0 >> RES 217 954 >> END >> Keep fixed with weak restraints >> 20.0
>> RES 1909 2646 >> END >> Keep fixed with weak restraints >> 20.0 >> RES 3601 4338 >> END >> res also >> 5.0 >> RES 955 1692 >> END >> res also >> 5.0 >> RES 2647 3384 >> END >> res also >> 5.0 >> RES 4339 5076 >> END
>> END >> >> >> >> Thanks again!!!!!!!!!!!!!!! >> >> >> >> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker <ross_at_rosswalker.co.uk> >> wrote: >> >>> Hi Wang, >>> >>> 800K atoms is pretty large and while sander / pmemd should support this >>> size >>> (I think
999,999 is the limit right now due to file formatting) you may >>> run >>> into problems that haven't been seen before. >>> >>> It's not obvious what is going wrong in your case but the numbers don't >>> make >>> any sense (a negative capacity!) which suggests either memory corruption >>> through an array overflow or the number of pairs is larger than a signed >>> integer and is overflowing. Even at 800K atoms you shouldn't have this
>>> many >>> pairs though. Can you post your input file so we can take a look? I >>> suspect >>> you have cut set too high or perhaps are not running PME etc. >>> >>> All the best >>> Ross >>> >>>> -----Original Message----- >>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On >>>> Behalf >>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM >>>> To: amber_at_scripps.edu >>>> Subject: AMBER: Non bond list error >>>> >>>> Hi, Dear AMBERs, >>>> >>>> I meet a problem when I run a simulation of a system consist of >>>> 799889 atoms. >>>> >>>> * NB pairs 451 0 exceeds capacity ( >>>> -28510921) 7
>>>> SIZE OF NONBOND LIST = -28510921 >>>> SANDER BOMB in subroutine nonbond_list >>>> Non bond list overflow! >>>> check MAXPR in locmem.f >>>> >>>> Could anyone tell me what's happen? >>>> >>>> Thanks a lot! >>>> >>>> ----------------------------------------------------------------------- >>>> The AMBER Mail Reflector >>>> To post,
send mail to amber_at_scripps.edu >>>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>>> to majordomo_at_scripps.edu >>> >>> ----------------------------------------------------------------------- >>> The AMBER Mail Reflector >>> To post, send mail to amber_at_scripps.edu >>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >>> to majordomo_at_scripps.edu
>>> >>> >> >> >> >> -- >> Wang,Ying >> >> ----------------------------------------------------------------------- >> The AMBER Mail Reflector >> To post, send mail to amber_at_scripps.edu >> To unsubscribe, send "unsubscribe amber" (in the *body* of the email) >> to majordomo_at_scripps.edu >>
> > ----------------------------------------------------------------------- > The AMBER Mail Reflector > To post, send mail to amber_at_scripps.edu > To unsubscribe, send "unsubscribe amber" (in the *body* of the email) > to majordomo_at_scripps.edu >
--- On Fri, 10/24/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 11:40 AM
Have you built pmemd? It is a separate step after the general build of amber. If you have built it, is it somewhere that it can be found? Here you just reference it as "pmemd" with no path. I generally use something like $AMBERHOME/exe/pmemd, ASSUMING it has been built...
Regards - Bob Duke
----- Original Message -----
From: neville forlemu
To: amber_at_scripps.edu
Sent: Friday, October 24, 2008 12:34 PM
Subject: RE: AMBER: Non bond list error
Hi Ross,
I did change the parameters you suggested to their defaults, and also changed the forcefield type, but got the same error from sanders.
However when I try pmemd this is the error I got
mpirun -np 4 pmemd -O -i minwat.in -o minwat.out -p rldhn_sol.top -crldhn_sol.crd -r minwat.resrst -ref rldhn_sol.crd
MPI: brinsvr1: 0x6c6bffff00005964: /usr/local/gnu/bin/bash: line 1: exec: pmemd: not found
MPI: could not run executable (all_signal.c:206)
Neville
--- On Fri, 10/24/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Friday, October 24, 2008, 1:12 AM
Hi Neville
You have a valid reason for this:
scnb = 1.0, scee = 1.0, ??????
You should NEVER EVER mess with the defaults of these values unless you know exactly what you are doing, i.e. using a non standard force field that uses different 1-4 scaling factors. Setting these to 1.0 for the AMBER FF series force fields will invalidate ALL of your results.
Note you also probably do not want to be using FF99. I suggest using FF99SB or FF03.
You should also not be messing with nsnb - leave that at the default as well.
As for the Non bond list overflow problem try setting cut=8.0 and see if it goes away.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 10:13 PM
To: amber_at_scripps.edu
Subject: RE: AMBER: Non bond list error
Hi Ross
I am using amber8 and trying to use sanders to minimize waters around this huge protein of about 20000 atoms, then perform a short md on some residues in the protein.
Constant Volume Minimization
# Control section
&cntrl
ntwx = 500, ntpr = 500, ntwr = 500,
scnb = 1.0, scee = 1.0, nsnb = 25, dielc = 1, cut = 12.0,
ntb = 1, ntc = 2, ntf = 2,
maxcyc = 10000, ntmin = 1, ncyc = 1000, drms = 0.0001,
ntp = 0,
ibelly = 0, ntr = 1,
imin = 1,
&end
Group Input for restrained atoms
5.0
RES 1 1332
END
END
Thanks
Neville
Also below is the xleap file preparation
Welcome to LEaP!
(no leaprc in search path)
> source leaprc.ff99
----- Source: /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99
----- Source of /opt/brinsoft/amber8/dat/leap/cmd/leaprc.ff99 done
Log file: ./leap.log
Loading parameters: /opt/brinsoft/amber8/dat/leap/parm/parm99.dat
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_nucleic94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_amino94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminoct94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/all_aminont94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/ions94.lib
Loading library: /opt/brinsoft/amber8/dat/leap/lib/solvents.lib
> x = loadpdb rldhN.pdb
Loading PDB file: ./rldhN.pdb
(starting new molecule for chain B)
(starting new molecule for chain C)
(starting new molecule for chain D)
total atoms in file: 10236
Leap added 10556 missing atoms according to residue templates:
10556 H / lone pairs
> savepdb x rld.pdb
Writing pdb file: rld.pdb
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
Shortening residue name for PDB format: NALA -> ALA
Shortening residue name for PDB format: CPHE -> PHE
> y = loadpdb rld.pdb
Loading PDB file: ./rld.pdb
total atoms in file: 20792
> addions y Cl- 0
8 Cl- ions required to neutralize.
Adding 8 counter ions to "y" using 1A grid
Grid extends from solute vdw + 2.47 to 8.47
Resolution: 1.00 Angstrom.
grid build: 1 sec
(no solvent present)
Calculating grid charges
charges: 251 sec
Placed Cl- in y at (5.12, -4.35, -51.18).
Placed Cl- in y at (4.12, 4.65, 0.82).
Placed Cl- in y at (13.12, -39.35, -35.18).
Placed Cl- in y at (-3.88, 16.65, -2.18).
Placed Cl- in y at (29.12, -0.35, -20.18).
Placed Cl- in y at (-8.88, -29.35, 3.82).
Placed Cl- in y at (-9.88, 32.65, -53.18).
Placed Cl- in y at (-27.88, -7.35, -31.18).
Done adding ions.
> solvateBox y TIP3PBOX 10
Solute vdw bounding box: 76.168 91.281 70.396
Total bounding box for atom centers: 96.168 111.281 90.396
Solvent unit box: 18.774 18.774 18.774
Total vdw box size: 99.135 92.364 120.398 angstroms.
Volume: 1102429.931 A^3
Total mass 702324.416 amu, Density 1.058 g/cc
> saveamberparm y rldhn_sol.top rldhn_sol.crd
Checking Unit.
Building topology.
Building atom parameters.
Building bond parameters.
Building angle parameters.
Building proper torsion parameters.
Building improper torsion parameters.
total 3824 improper torsions applied
Building H-Bond parameters.
Not Marking per-residue atom chain types.
Marking per-residue atom chain types.
(Residues lacking connect0/connect1 -
these don't have chain types marked:
--- On Thu, 10/23/08, Ross Walker <ross_at_rosswalker.co.uk> wrote:
From: Ross Walker <ross_at_rosswalker.co.uk>
Subject: RE: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 11:56 PM
Hi Neville,
Can you post your input file and some details about your system (number of atoms, type of simulation, gas phase, periodic boundaries etc) as well as the version of AMBER you are using.
All the best
Ross
From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of neville forlemu
Sent: Thursday, October 23, 2008 9:48 PM
To: amber_at_scripps.edu; rduke_at_email.unc.edu
Subject: Re: AMBER: Non bond list error
Hello,
Could some one explain to me what this error means
* NB pairs 7104 103754005 exceeds capacity ( 103754298) 0
SIZE OF NONBOND LIST = 103754298
SANDER BOMB in subroutine nonbond_list
Non bond list overflow!
check MAXPR in locmem.f
I am trying to run sander for energy minimization, but keep running into this problem.
Thanks
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show
that
you didn't... (oh, oops). So it is memory corruption. What I don't
understand is why you are not dying with some sort of "out of memory"
error
from sander, associated with asking for more memory than is available. For
pmemd, anywhere I allocate dynamic memory, I check for a success return
code, so the way you should experience running out of memory there is to get
an explicit error message. Because sander has a preallocated memory pool
strategy, I suspect that other things are possible... Bottom line on all
this - I think it is a good idea to not run more than roughly 100,000 atoms
on a single processor, especially for sander. And if you run it on 4
processors but they all share the same limited physical memory, you may also
hit trouble. I attached a graphic on pmemd memory requirements - a jpg so
it should be widely viewable. My rule of thumb for pmemd is that 4
processors, each with 1 GB of actual physical memory, can handle up to 1
million atoms with the default 8 angstrom cutoff. Sander will take more.
There are also buffer space considerations in an mpi application (within mpi
itself, not in the app), that further muddy the waters, but following this
guideline you should be safe.
Regards - Bob
----- Original Message -----
From: "Robert Duke" <rduke_at_email.unc.edu>
To: <amber_at_scripps.edu>
Sent: Thursday, October 23, 2008 4:14 PM
Subject: Re: AMBER: Non bond list error
> As Ross will tell you too:
> 1) Don't increase cut to 12, leave it at the default (of 8)
> 2) Run this on at least 4 processors using the MPI version of pmemd or
> sander (I know you are using sander here; pmemd requires less memory).
> Even higher processor counts will reduce your risk of memory overflow
> further. Your pairlist went negative because you incremented it past a 31
> bit digit; with the commonly used integer format on computers these days
> (twos-complement), this results in a negative number (and is clearly an
> error condition). Is this memory usage reasonable for the size problem
> you have? Well, that cutoff plus skin will produce about 552 pairs per
> atom. If you had 1,000,000 atoms (and you are close), that would be
> 552,000,000 pairs. Not enough to overflow the list counter. BUT that is
> 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list
> alone. Most machines, you are pushing it to get much over 1.5 GB for the
> application (I have not looked recently, so that is off the top of my
> head). With true 32 bit executables, you are out of address space; with
> the newer 64 bit chips, you have bits to specify more than 2 GB of
> addresses, but you may not have enough actual memory. And remember that
> the pairlist is only part of your memory consumption. No resource is
> infinite on a computer...
> Regards - Bob Duke
> ----- Original Message -----
> From: "Wang,Ying" <wangying_at_ufl.edu>
> To: <amber_at_scripps.edu>
> Sent: Thursday, October 23, 2008 3:26 PM
> Subject: RE: AMBER: Non bond list error
>
>
>> Hi, Ross,
>>
>> Thanks a lot!
>>
>> My input file is as below:
>> 50ps MD with res
>> &cntrl
>> imin = 0,
>> irest = 0,
>> ntx = 1,
>> ntb = 1,
>> cut = 12,
>> ntr = 1,
>> ntc = 2,
>> ntf = 2,
>> tempi = 0.0,
>> temp0 = 300.0,
>> ntt = 3,
>> gamma_ln = 2.0,
>> nstlim = 50000, dt = 0.001
>> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> nmropt=1
>> /
>> &wt TYPE='TEMP0', istep1=0, istep2=50000,
>> value1=0.1, value2=300.0, /
>> &wt TYPE='END' /
>> Keep system fixed with weak restraints
>> 20.0
>> RES 1 5076
>> END
>> END
>>
>> and the NPT is as below:
>>
>> NPT: 50ps MD
>> &cntrl
>> imin = 0, irest = 1, ntx = 7,
>> ntb = 2, pres0 = 1.0, ntp = 1,
>> taup = 2.0,
>> cut = 12, ntr = 1,
>> ntc = 2, ntf = 2,
>> tempi = 300.0, temp0 = 300.0,
>> ntt = 3, gamma_ln = 2.0,
>> nstlim = 50000, dt = 0.001,
>> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> /
>> Keep fixed with weak restraints
>> 20.0
>> RES 217 954
>> END
>> Keep fixed with weak restraints
>> 20.0
>> RES 1909 2646
>> END
>> Keep fixed with weak restraints
>> 20.0
>> RES 3601 4338
>> END
>> res also
>> 5.0
>> RES 955 1692
>> END
>> res also
>> 5.0
>> RES 2647 3384
>> END
>> res also
>> 5.0
>> RES 4339 5076
>> END
>> END
>>
>>
>>
>> Thanks again!!!!!!!!!!!!!!!
>>
>>
>>
>> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker
<ross_at_rosswalker.co.uk>
>> wrote:
>>
>>> Hi Wang,
>>>
>>> 800K atoms is pretty large and while sander / pmemd should support
this
>>> size
>>> (I think 999,999 is the limit right now due to file formatting)
you may
>>> run
>>> into problems that haven't been seen before.
>>>
>>> It's not obvious what is going wrong in your case but the
numbers don't
>>> make
>>> any sense (a negative capacity!) which suggests either memory
corruption
>>> through an array overflow or the number of pairs is larger than a
signed
>>> integer and is overflowing. Even at 800K atoms you shouldn't
have this
>>> many
>>> pairs though. Can you post your input file so we can take a look?
I
>>> suspect
>>> you have cut set too high or perhaps are not running PME etc.
>>>
>>> All the best
>>> Ross
>>>
>>>> -----Original Message-----
>>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu]
On
>>>> Behalf
>>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM
>>>> To: amber_at_scripps.edu
>>>> Subject: AMBER: Non bond list error
>>>>
>>>> Hi, Dear AMBERs,
>>>>
>>>> I meet a problem when I run a simulation of a system consist
of
>>>> 799889 atoms.
>>>>
>>>> * NB pairs 451 0 exceeds capacity (
>>>> -28510921) 7
>>>> SIZE OF NONBOND LIST = -28510921
>>>> SANDER BOMB in subroutine nonbond_list
>>>> Non bond list overflow!
>>>> check MAXPR in locmem.f
>>>>
>>>> Could anyone tell me what's happen?
>>>>
>>>> Thanks a lot!
>>>>
>>>>
-----------------------------------------------------------------------
>>>> The AMBER Mail Reflector
>>>> To post, send mail to amber_at_scripps.edu
>>>> To unsubscribe, send "unsubscribe amber" (in the
*body* of the email)
>>>> to majordomo_at_scripps.edu
>>>
>>>
-----------------------------------------------------------------------
>>> The AMBER Mail Reflector
>>> To post, send mail to amber_at_scripps.edu
>>> To unsubscribe, send "unsubscribe amber" (in the *body*
of the email)
>>> to majordomo_at_scripps.edu
>>>
>>>
>>
>>
>>
>> --
>> Wang,Ying
>>
>>
-----------------------------------------------------------------------
>> The AMBER Mail Reflector
>> To post, send mail to amber_at_scripps.edu
>> To unsubscribe, send "unsubscribe amber" (in the *body* of
the email)
>> to majordomo_at_scripps.edu
>>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the
email)
> to majordomo_at_scripps.edu
>
--- On Thu, 10/23/08, Robert Duke <rduke_at_email.unc.edu> wrote:
From: Robert Duke <rduke_at_email.unc.edu>
Subject: Re: AMBER: Non bond list error
To: amber_at_scripps.edu
Date: Thursday, October 23, 2008, 3:44 PM
Actually, I noticed I said "you overflowed the counter", and then show
that
you didn't... (oh, oops). So it is memory corruption. What I don't
understand is why you are not dying with some sort of "out of memory"
error
from sander, associated with asking for more memory than is available. For
pmemd, anywhere I allocate dynamic memory, I check for a success return
code, so the way you should experience running out of memory there is to get
an explicit error message. Because sander has a preallocated memory pool
strategy, I suspect that other things are possible... Bottom line on all
this - I think it is a good idea to not run more than roughly 100,000 atoms
on a single processor, especially for sander. And if you run it on 4
processors but they all share the same limited physical memory, you may also
hit trouble. I attached a graphic on pmemd memory requirements - a jpg so
it should be widely viewable. My rule of thumb for pmemd is that 4
processors, each with 1 GB of actual physical memory, can handle up to 1
million atoms with the default 8 angstrom cutoff. Sander will take more.
There are also buffer space considerations in an mpi application (within mpi
itself, not in the app), that further muddy the waters, but following this
guideline you should be safe.
Regards - Bob
----- Original Message -----
From: "Robert Duke" <rduke_at_email.unc.edu>
To: <amber_at_scripps.edu>
Sent: Thursday, October 23, 2008 4:14 PM
Subject: Re: AMBER: Non bond list error
> As Ross will tell you too:
> 1) Don't increase cut to 12, leave it at the default (of 8)
> 2) Run this on at least 4 processors using the MPI version of pmemd or
> sander (I know you are using sander here; pmemd requires less memory).
> Even higher processor counts will reduce your risk of memory overflow
> further. Your pairlist went negative because you incremented it past a 31
> bit digit; with the commonly used integer format on computers these days
> (twos-complement), this results in a negative number (and is clearly an
> error condition). Is this memory usage reasonable for the size problem
> you have? Well, that cutoff plus skin will produce about 552 pairs per
> atom. If you had 1,000,000 atoms (and you are close), that would be
> 552,000,000 pairs. Not enough to overflow the list counter. BUT that is
> 552,000,000 pairs * 4 bytes per integer, means 2 GB in the nonbonded list
> alone. Most machines, you are pushing it to get much over 1.5 GB for the
> application (I have not looked recently, so that is off the top of my
> head). With true 32 bit executables, you are out of address space; with
> the newer 64 bit chips, you have bits to specify more than 2 GB of
> addresses, but you may not have enough actual memory. And remember that
> the pairlist is only part of your memory consumption. No resource is
> infinite on a computer...
> Regards - Bob Duke
> ----- Original Message -----
> From: "Wang,Ying" <wangying_at_ufl.edu>
> To: <amber_at_scripps.edu>
> Sent: Thursday, October 23, 2008 3:26 PM
> Subject: RE: AMBER: Non bond list error
>
>
>> Hi, Ross,
>>
>> Thanks a lot!
>>
>> My input file is as below:
>> 50ps MD with res
>> &cntrl
>> imin = 0,
>> irest = 0,
>> ntx = 1,
>> ntb = 1,
>> cut = 12,
>> ntr = 1,
>> ntc = 2,
>> ntf = 2,
>> tempi = 0.0,
>> temp0 = 300.0,
>> ntt = 3,
>> gamma_ln = 2.0,
>> nstlim = 50000, dt = 0.001
>> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> nmropt=1
>> /
>> &wt TYPE='TEMP0', istep1=0, istep2=50000,
>> value1=0.1, value2=300.0, /
>> &wt TYPE='END' /
>> Keep system fixed with weak restraints
>> 20.0
>> RES 1 5076
>> END
>> END
>>
>> and the NPT is as below:
>>
>> NPT: 50ps MD
>> &cntrl
>> imin = 0, irest = 1, ntx = 7,
>> ntb = 2, pres0 = 1.0, ntp = 1,
>> taup = 2.0,
>> cut = 12, ntr = 1,
>> ntc = 2, ntf = 2,
>> tempi = 300.0, temp0 = 300.0,
>> ntt = 3, gamma_ln = 2.0,
>> nstlim = 50000, dt = 0.001,
>> ntpr = 1000, ntwx = 1000, ntwr = 1000
>> /
>> Keep fixed with weak restraints
>> 20.0
>> RES 217 954
>> END
>> Keep fixed with weak restraints
>> 20.0
>> RES 1909 2646
>> END
>> Keep fixed with weak restraints
>> 20.0
>> RES 3601 4338
>> END
>> res also
>> 5.0
>> RES 955 1692
>> END
>> res also
>> 5.0
>> RES 2647 3384
>> END
>> res also
>> 5.0
>> RES 4339 5076
>> END
>> END
>>
>>
>>
>> Thanks again!!!!!!!!!!!!!!!
>>
>>
>>
>> On Thu Oct 23 14:38:09 EDT 2008, Ross Walker
<ross_at_rosswalker.co.uk>
>> wrote:
>>
>>> Hi Wang,
>>>
>>> 800K atoms is pretty large and while sander / pmemd should support
this
>>> size
>>> (I think 999,999 is the limit right now due to file formatting)
you may
>>> run
>>> into problems that haven't been seen before.
>>>
>>> It's not obvious what is going wrong in your case but the
numbers don't
>>> make
>>> any sense (a negative capacity!) which suggests either memory
corruption
>>> through an array overflow or the number of pairs is larger than a
signed
>>> integer and is overflowing. Even at 800K atoms you shouldn't
have this
>>> many
>>> pairs though. Can you post your input file so we can take a look?
I
>>> suspect
>>> you have cut set too high or perhaps are not running PME etc.
>>>
>>> All the best
>>> Ross
>>>
>>>> -----Original Message-----
>>>> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu]
On
>>>> Behalf
>>>> Of Wang,Ying
>>>> Sent: Thursday, October 23, 2008 10:20 AM
>>>> To: amber_at_scripps.edu
>>>> Subject: AMBER: Non bond list error
>>>>
>>>> Hi, Dear AMBERs,
>>>>
>>>> I meet a problem when I run a simulation of a system consist
of
>>>> 799889 atoms.
>>>>
>>>> * NB pairs 451 0 exceeds capacity (
>>>> -28510921) 7
>>>> SIZE OF NONBOND LIST = -28510921
>>>> SANDER BOMB in subroutine nonbond_list
>>>> Non bond list overflow!
>>>> check MAXPR in locmem.f
>>>>
>>>> Could anyone tell me what's happen?
>>>>
>>>> Thanks a lot!
>>>>
>>>>
-----------------------------------------------------------------------
>>>> The AMBER Mail Reflector
>>>> To post, send mail to amber_at_scripps.edu
>>>> To unsubscribe, send "unsubscribe amber" (in the
*body* of the email)
>>>> to majordomo_at_scripps.edu
>>>
>>>
-----------------------------------------------------------------------
>>> The AMBER Mail Reflector
>>> To post, send mail to amber_at_scripps.edu
>>> To unsubscribe, send "unsubscribe amber" (in the *body*
of the email)
>>> to majordomo_at_scripps.edu
>>>
>>>
>>
>>
>>
>> --
>> Wang,Ying
>>
>>
-----------------------------------------------------------------------
>> The AMBER Mail Reflector
>> To post, send mail to amber_at_scripps.edu
>> To unsubscribe, send "unsubscribe amber" (in the *body* of
the email)
>> to majordomo_at_scripps.edu
>>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the
email)
> to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
to majordomo_at_scripps.edu
- Next message: Robert Duke: "Re: AMBER: Non bond list error"
- Previous message: Francesco Pietra: "AMBER: Atom names CL and four characters"
- In reply to: neville forlemu: "Re: AMBER: Non bond list error"
- Next in thread: Robert Duke: "Re: AMBER: Non bond list error"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
|
VU Home |
VUMC Home |
People Finder |
University Calendar
webmaster- modified on January 30, 2009 |