VMD-L Mailing List

From: Peter Freddolino (petefred_at_umich.edu)
Date: Thu Mar 07 2019 - 22:42:30 CST

I can't really comment without more details about the exact series of
commands that you're attempting, but my guess would be (1) yes, (2) are you
actually getting useful data out of running 30,000 volmap frames in light
of the long autocorrelation times that may be at play in your system, and
(3) if so, could you just calculate the volmaps on chunks of the trajectory
and then average or otherwise summarize them afterwards?

Best,
Peter

On Thu, Mar 7, 2019 at 10:13 PM McGuire, Kelly <mcg05004_at_byui.edu> wrote:

> Hey Peter, I will give that a try. I also notice another potential memory
> issue with volmap. I can't get volmap to work beyond 30,000 frames without
> VMD crashing. I wonder if the BigDCD plugin would fix this memory issue as
> well...?
>
> *Kelly L. McGuire*
>
> *PhD Candidate*
>
> *Biophysics*
>
> *Department of Physiology and Developmental Biology*
>
> *Brigham Young University*
>
> *LSB 3050*
>
> *Provo, UT 84602*
>
>
> ------------------------------
> *From:* Peter Freddolino <petefred_at_umich.edu>
> *Sent:* Thursday, March 7, 2019 7:58 PM
> *To:* McGuire, Kelly
> *Cc:* Bennion, Brian; VMD Mailing LIst
> *Subject:* Re: vmd-l: Re: VMD Text Mode
>
> Short answer is yes, you're still going to be memory limited. Possible
> solutions are loading only one every n frames, doing some analysis on
> trajectories with water removed, or using bigdcd (
> https://www.ks.uiuc.edu/Research/vmd/script_library/scripts/bigdcd/).
> BTW, you may want to get into the habit of adding 'waitfor all' to the end
> of your load commands if you're working with big trajectories. Otherwise
> the tcl interpreter can move on after the load starts but before it
> finishes, which can also lead to odd looking results downstream.
> Best,
> Peter
>
> On Thu, Mar 7, 2019 at 9:53 PM McGuire, Kelly <mcg05004_at_byui.edu> wrote:
>
> I tried this on Linux, RHEL7 and Ubuntu. RHEL7 is on the supercomputer
> and has 123 GB RAM available on the system I use for VMD. I was able to
> open 38,000 frames on RHEL7, but couldn't get the full 90,000. My DCD is
> 75 GB in size.
>
> Ubuntu was local, so it had 32 GB RAM available. I was only able to get
> 3,200 frames loaded in text mode.
>
>
> VMD does not crash, it just loads a reduced number of frames.
>
>
> *Kelly L. McGuire*
>
> *PhD Candidate*
>
> *Biophysics*
>
> *Department of Physiology and Developmental Biology*
>
> *Brigham Young University*
>
> *LSB 3050*
>
> *Provo, UT 84602*
>
>
> ------------------------------
> *From:* Bennion, Brian <bennion1_at_llnl.gov>
> *Sent:* Thursday, March 7, 2019 6:07 PM
> *To:* McGuire, Kelly; VMD Mailing LIst
> *Subject:* RE: VMD Text Mode
>
>
> Does the vmd session crash? What is your hardware/OS that you are using
> for the textmode session?
>
>
>
> Brian
>
>
>
> *From:* owner-vmd-l_at_ks.uiuc.edu <owner-vmd-l_at_ks.uiuc.edu> *On Behalf Of *McGuire,
> Kelly
> *Sent:* Thursday, March 7, 2019 4:02 PM
> *To:* VMD Mailing LIst <vmd-l_at_ks.uiuc.edu>
> *Subject:* vmd-l: VMD Text Mode
>
>
>
> I am trying VMD text mode for the first time since I have really large DCD
> files. I tried loading my DCD, but it doesn't seem to load all of the
> frames. I do:
>
>
>
> mol new Simulation_QwikMD.psf type psf
>
> mol addfile MD1.dcd type dcd first 0 last -1 step 1
>
>
>
> There should be 90,000 frames, but i only loads ~3,200 frames. What am I
> doing wrong? Is VMD text mode still RAM dependent for DCD file sizes that
> can be loaded?
>
>
>
>
>
>
>
> *Kelly L. McGuire*
>
> *PhD Candidate*
>
> *Biophysics*
>
> *Department of Physiology and Developmental Biology*
>
> *Brigham Young University*
>
> *LSB 3050*
>
> *Provo, UT 84602*
>
>
>
>