Re: amber-developers: -suffix option for sander?

From: Robert Duke <rduke.email.unc.edu>
Date: Thu, 9 Feb 2006 20:25:11 -0700

Tom -
Further note. As specified in the original description, not only does the

suffix not apply to input; it does not apply to any file where you specify

the name explicitly (really just a note on added flexibility). But a
rewrite of your example becomes:

set prev = 1;
set cur = 2;

sander -O -c restrt.$prev -suffix $cur

and then update the script to increment $pre and $cur.

Not a huge deal I know, but it does work well into the sequence of runs
requirement, and creates a compact and cohesive command line syntax. For
me, I get away with bench and test of the form:

pmemd -O -suffix foo

- and I'm done...

Regards - Bob

----- Original Message -----
From: "Robert Duke" <rduke.email.unc.edu>
To: <amber-developers.scripps.edu>
Sent: Thursday, February 09, 2006 10:09 PM
Subject: Re: amber-developers: -suffix option for sander?


> Tom -
> The suffix only applies to output files, not mdin, prmtop, inpcrd or
refc.
> I find this sort of thing useful in linking all the associated output
> together, though obviously you can solve the problem via scripting too.
> However, with -suffix, then you can do it all in two spots - you
increment
> the suffix, and apply the last suffix value to your restrt.
Furthermore,
> you don't have to keep track of which outputs you actually requested;
they
> will all get tagged the same way. It is not a huge deal, but I put it
in
> pmemd because it saves me a ton of typing on the ~20 different systems I

> am running on testing, benchmarking, and occasionally even collecting
> data.
> Regards - Bob
>
> ----- Original Message -----
> From: "Thomas Cheatham" <cheatham.chpc.utah.edu>
> To: <amber-developers.scripps.edu>
> Sent: Thursday, February 09, 2006 9:58 PM
> Subject: Re: amber-developers: -suffix option for sander?
>
>
>>
>>> > 4) a new -suffix command line option. If you put -suffix
>>> > 020906.1.4proc
>>> > in the command line, then all the output files will have this
appended
>>> > to
>>> > the default name unless you explicitly specified a name for them on
>>> > the
>>> > command line. So this applies to mdout, mdinfo, mden, mdcrd, mdvel,
>>> > restrt, and logfile (pmemd now permits logfile naming also with a -l

>>> > flag).
>>> > This is a really simple, easy-to-use command line interface
>>> > enhancement you
>>> > all may want to pick up for sander. The code is in get_cmdline.fpp
>>> > for
>>> > pmemd.
>>>
>>> It sounds like a nice idea to me--it would get us a long way towards
>>> having
>>> a more standardized file nomenclature. I'm just posting this to the
>>> developers' list to see if people have either objections or better
ideas
>>> along
>>> this line.
>>
>> I like this idea in some respect, but most likely would not use it in
>> general. As I run now, I do:
>>
>> set pre = 1;
>> set cur = 2;
>>
>> sander -O -c restrt.$pre -o mdout.$cur -i mdin -p prmtop -r
>> restrt.$cur -x traj.$cur
>>
>> and then update the script to increment $pre and $cur.
>>
>> The -suffix option doesn't help me in this case since (1) both the
prmtop
>> and mdin files do not need a suffix as the same files are typically
used
>> throughout a long series of runs (i.e. I don't want to have to move the
>> mdin and prmtop files to prmtop.suffix/mdin.suffix every run as I
update
>> suffix), and (2) I would have to move the previous restrt to the new
>> inpcrd.suffix; so, I do not see a significant benefit for my usage
>> pattern.
>>
>> I would happily use -prefix though! This would be advantageous for
>> multisander since I could for each instance use the default names and
>> give
>> each a different sub-directory for the files via -prefix ...
>>
>> --tom
>>
>> (p.s. DAC, ptraj in the tree was -DBINTRAJ wrapped and I'll get on
final
>> NetCDF mods, cluster + docs next week)
>>
>>
>
>
>
Received on Wed Apr 05 2006 - 23:49:43 PDT
Custom Search