AW: AW: compress PSF into INTER

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Thu Jun 07 2012 - 08:25:10 CDT

As far as I understood, running namd with the genCompressedPSF should create
a new file within the same directory as the uncompressed psf that can be
used with useComnpressedPSF. Are there no new files? Try "dir -t" to sort by
date.

Norman Geist.

> -----Ursprüngliche Nachricht-----
> Von: Cruz-Chu Eduardo Roberto [mailto:eduardo.cruz_at_mavt.ethz.ch]
> Gesendet: Donnerstag, 7. Juni 2012 14:03
> An: Namd Mailing List
> Cc: JC Gumbart; Norman Geist; char_at_ks.uiuc.edu
> Betreff: RE: AW: namd-l: compress PSF into INTER
>
> Hi;
>
> I am still trying to generate the INTER file. That is, just the first
> step before running anything, only switching the structure file using
> "genCompressedPsf on".
>
> Not need of "useCompressedPsf on" yet, just the INTER file.
>
> So I moved to a small system of about 9000 atoms, just to check the
> thing is working. Unfortunately, I was not able to obtain the INTER
> file.
>
> I did three attemps:
>
> 1.- NAMD, binary downloaded from website.
> 2.- NAMD, normal compilation.
> 3.- NAMD, compiled with "-with-memopt" flag.
>
> The log files are attached below, and also the configuration file.
>
> If you had done it before (Chris?), please just send me your namd
> config file, even outside the list, so I can see what I am doing wrong.
>
> Regards;
>
> Eduardo
>
>
> ________________________________________
> From: owner-namd-l_at_ks.uiuc.edu [owner-namd-l_at_ks.uiuc.edu] on behalf of
> JC Gumbart [gumbart_at_ks.uiuc.edu]
> Sent: Thursday, June 07, 2012 12:26 PM
> To: Norman Geist
> Cc: Namd Mailing List
> Subject: Re: AW: namd-l: compress PSF into INTER
>
> I have generated compressed psfs before and can assure you that there
> should be more in the log file than what Eduardo listed. I also trust
> that he would have checked his directory for additional files after
> exit.
>
>
> On Jun 7, 2012, at 5:15 AM, Norman Geist wrote:
>
> >> -----Ursprüngliche Nachricht-----
> >> Von: Norman Geist [mailto:norman.geist_at_uni-greifswald.de]
> >> Gesendet: Donnerstag, 7. Juni 2012 12:12
> >> An: 'Cruz-Chu Eduardo Roberto'
> >> Cc: Namd Mailing List (namd-l_at_ks.uiuc.edu)
> >> Betreff: AW: namd-l: compress PSF into INTER
> >>
> >> Hi,
> >>
> >> I'm sorry to tell you that you haven't followed the tutorial very
> >> carefully. The keyword "genCompressedPsf on" does only generate a
> new
> >> compressed psf when running namd and than exits. To use the new
> >> compressed psf afterwards, you need to specify "useCompressedPsf on"
> >> and remove the "genCompressedPsf on" from the conf. Also use
> >> structure.inter instead of structure in the conf. Also you should
> use
> >> bincoordinates instead of coordinates which means that you have to
> >> convert your non binary pdb or coor to a binary one (using vmd).
> >>
> >> So the bottom line is it's two steps:
> >>
> >> 1. Generate a compressed psf with namd and genCompressedPsf on" and
> a
> >> binary coordinate file with vmd.
> >> 2. Run memopt namd with "useCompressedPsf on"
> >
> > and structure.inter
> >
> > I forgot to mention.
> >
> >>
> >> I hope I got it right.
> >>
> >> Good luck.
> >>
> >> For more information look at
> >> http://www.ks.uiuc.edu/Research/namd/wiki/?NamdMemoryReduction
> >>
> >> Norman Geist.
> >>
> >>
> >>> -----Ursprüngliche Nachricht-----
> >>> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> >>> Auftrag von Cruz-Chu Eduardo Roberto
> >>> Gesendet: Donnerstag, 7. Juni 2012 11:21
> >>> An: Norman Geist
> >>> Cc: Namd Mailing List
> >>> Betreff: RE: namd-l: compress PSF into INTER
> >>>
> >>> Hi Norman;
> >>>
> >>> NAMD just halts after reading the PSF file, then NAMD crashes. The
> >>> last line of the log file are:
> >>>
> >>> Info: 0 VDW_PAIRS
> >>> Info: 0 NBTHOLE_PAIRS
> >>> Info: TIME FOR READING PSF FILE: 34.5907
> >>>
> >>> My desktop has 16 GB of memory, so it should handle the PSF/COOR
> >> files,
> >>> which are less than 2GB of data.
> >>>
> >>> Below I am attaching the log file.
> >>>
> >>> Have you (or anyone) successfully used the "genCompressedPsf on"
> >> option
> >>> to generate an INTER file?. I pay a beer for any positive answer.
> >>>
> >>> Regards;
> >>>
> >>> Eduardo
> >>> ________________________________________
> >>> From: Norman Geist [norman.geist_at_uni-greifswald.de]
> >>> Sent: Thursday, June 07, 2012 8:24 AM
> >>> To: Cruz-Chu Eduardo Roberto
> >>> Cc: Namd Mailing List
> >>> Subject: AW: namd-l: compress PSF into INTER
> >>>
> >>>> -----Ursprüngliche Nachricht-----
> >>>> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> >>>> Auftrag von Cruz-Chu Eduardo Roberto
> >>>> Gesendet: Mittwoch, 6. Juni 2012 20:38
> >>>> An: namd-l_at_ks.uiuc.edu
> >>>> Betreff: namd-l: compress PSF into INTER
> >>>>
> >>>> Hi;
> >>>>
> >>>
> >>> Hi
> >>>
> >>>>
> >>>> I am trying to run a 6 million atom system. For that, I need to
> >>>
> >>> Wow, that's heavy. My biggest system has 1.5 million ^^
> >>>
> >>>> compress the PSF file into a INTER file. The NAMD wiki provide
> >> some
> >>>> brief explanation.
> >>>>
> >>>>
> >>>>
> >>>
> >>
> http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdMemoryReducti
> >> o
> >>>> n
> >>>>
> >>>>
> >>>> that is "b) add a field "genCompressedPsf on" in the configuration
> >>>> file".
> >>>>
> >>>>
> >>>> well, I did it using NAMD 2.8 and also NAMD 2.9. I also compiled
> >> NAMD
> >>>> with --with-memopt flag, and no success.
> >>>
> >>> What exactly does "no success" mean? Is namd crashing with an error
> >> or
> >>> something?
> >>> We need some detail to identify the problem.
> >>>
> >>> Regards
> >>>
> >>> Norman
> >>>
> >>>>
> >>>>
> >>>> Does anyone has an example?, a namd configuration file I can use
> as
> >>>> guide?.
> >>>>
> >>>>
> >>>> Below, I am attaching my namd configuration, the one that did not
> >>> work.
> >>>>
> >>>>
> >>>> Any help, it would be really appreciated.
> >>>>
> >>>>
> >>>> Regards;
> >>>>
> >>>>
> >>>> Eduardo
> >>>>
> >>>>
> >>>
> >
> >
>

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:22:05 CST